## Time-stamp: <Sat Jun 28 23:22:06 2014 Ashton Trey Belew (abelew@gmail.com)>
load("RData")
source("R/myr.R")
##rm(list=ls())
##save(list = ls(all=TRUE), file="RData")Genome annotation input
I am pulling data from 3 primary sources: * The gas gff annotations/fasta genome * The NCBI reference genome for the strain 5005 * The microbesonline.org tab delimited file.
More information may be found in the reference/ directory.
annotations = import.gff3("reference/gff/MGAS_5005.gff.gz", asRangedData=FALSE)
annotation_info = as.data.frame(annotations)
genes = annotation_info[annotation_info$type=="gene",]
rownames(genes) = genes$locus_tag
gene_annotations = subset(genes, select = c("start", "end", "width", "strand", "Name", "protein_id", "ID"))
short_annotations = gene_annotations[,c("ID", "Name")]
short_annotations$spy = rownames(short_annotations)
colnames(short_annotations) = c("ID", "Name", "Spy")
write_xls(genes, "genes", rowname="ID")
microbes = read.csv(file="reference/microbesonline/MGAS_5005_annotations.tab.gz", header=1, sep="\t")
microbes_go = microbes[,c("sysName","GO")]
go_entries = strsplit(as.character(microbes_go$GO), split=",", perl=TRUE)
microbes_go_oneperrow = data.frame(name = rep(microbes_go$sysName, sapply(go_entries, length)), GO = unlist(go_entries))
microbes_go = microbes_go_oneperrow
rm(microbes_go_oneperrow)
rm(go_entries)
## These are used for gene ontology stuff...
microbes_lengths = microbes[,c("sysName", "start","stop")]
microbes_lengths$length = abs(microbes$start - microbes$stop)
microbes_lengths = microbes_lengths[,c("sysName","length")]Make tooltips for interactive graphs
I am going to be potentially throwing tooltips all over the place, so let us make a quick datastructure for them.
tooltip_data = genes
tooltip_data = tooltip_data[,c("ID","Name", "locus_tag")]
tooltip_data$tooltip = paste(tooltip_data$Name, tooltip_data$locus_tag, sep=": ")
tooltip_data$tooltip = gsub("\\+", " ", tooltip_data$tooltip)
rownames(tooltip_data) = tooltip_data$ID
tooltip_data = tooltip_data[-1]
tooltip_data = tooltip_data[-1]
tooltip_data = tooltip_data[-1]
head(tooltip_data)## tooltip
## gene0 dnaA: M5005_Spy_0001
## gene1 dnaN: M5005_Spy_0002
## gene2 M5005_Spy_0003: M5005_Spy_0003
## gene3 ychF: M5005_Spy_0004
## gene4 pth: M5005_Spy_0005
## gene5 trcF: M5005_Spy_0006Set up the experimental design
I have a sample design csv file called all_samples.csv which should contain all the needed information for this process.
The count tables come from the preprocessing directory. In each HPGL directory there is a counts directory. I am pulling from these the count tables generated from bowtie runs which asked for a randomly assigned multimatch (-M 1) with 0 mismatches (-v 0). These files are called something like: 05v0M1gen_NZ131_t0_genome.count as an example.
library_colors = list(lib1="yellow4",
lib2="darkgreen",
lib3="red",
lib4="darkblue")
time_colors = list(t0="lightgray",
t1="darkgray",
t2="black")
## The all_samples.csv controls the experimental settings
## all the following lines manipulate the information therein
## in order to set up media types, replicates, etc
sample_definitions = read.csv(file="all_samples_5448.csv", sep=",")
## If I have summary lines in the csv, they will not start with 'HPGL' and so should be dropped.
sample_definitions = sample_definitions[grepl('^HPGL', sample_definitions$Sample.ID, perl=TRUE),]
## Pre set the color scheme
sample_definitions$colors = library_colors
## This long statement just writes out a computer path name containing the count files to read by sample name.
sample_definitions$counts = paste("data/count_tables/", sample_definitions$Strain, "-", sample_definitions$Replicate, sample_definitions$Time, ".count.gz", sep="")
sample_definitions = as.data.frame(sample_definitions)
rownames(sample_definitions) = sample_definitions$Sample.ID
## my_read_files does just that, it reads the count tables and makes a large raw count table from them.
all_count_tables = my_read_files(as.character(sample_definitions$Sample.ID), as.character(sample_definitions$counts))
## make it into a matrix for use as an expressionset
all_count_matrix = as.matrix(all_count_tables)
gene_info = all_count_matrix[rownames(all_count_matrix) %in% genes$ID,]
all_count_matrix = all_count_matrix[rownames(all_count_matrix) %in% genes$ID,]
metadata = new("AnnotatedDataFrame", data.frame(sample=sample_definitions$Sample.ID,
condition=sample_definitions$Time,
batch=sample_definitions$Library,
strain=sample_definitions$Strain,
color=as.character(sample_definitions$colors),
counts=sample_definitions$counts))
## Now generate the expressionset object
sampleNames(metadata) = colnames(all_count_matrix)
feature_data = new("AnnotatedDataFrame", as.data.frame(gene_info))
featureNames(feature_data) = rownames(all_count_matrix)
experiment = new("ExpressionSet", exprs=all_count_matrix,
phenoData=metadata, featureData=feature_data)
## print some information to see that it worked
print(experiment)## ExpressionSet (storageMode: lockedEnvironment)
## assayData: 1950 features, 16 samples
## element names: exprs
## protocolData: none
## phenoData
## sampleNames: HPGL0259 HPGL0260 ... HPGL0274 (16 total)
## varLabels: sample condition ... counts (6 total)
## varMetadata: labelDescription
## featureData
## featureNames: gene0 gene1 ... gene999 (1950 total)
## fvarLabels: HPGL0259 HPGL0260 ... HPGL0274 (16 total)
## fvarMetadata: labelDescription
## experimentData: use 'experimentData(object)'
## Annotation:summary(exprs(experiment))## HPGL0259 HPGL0260 HPGL0261 HPGL0262
## Min. : 0 Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 76 1st Qu.: 136 1st Qu.: 38 1st Qu.: 4
## Median : 328 Median : 694 Median : 165 Median : 20
## Mean : 680 Mean : 2528 Mean : 2438 Mean : 1151
## 3rd Qu.: 760 3rd Qu.: 1985 3rd Qu.: 487 3rd Qu.: 93
## Max. :157502 Max. :1285417 Max. :1926727 Max. :1300741
## HPGL0263 HPGL0264 HPGL0265 HPGL0266
## Min. : 0 Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 8 1st Qu.: 6 1st Qu.: 5 1st Qu.: 8
## Median : 39 Median : 29 Median : 19 Median : 31
## Mean : 1797 Mean : 1807 Mean : 1759 Mean : 1688
## 3rd Qu.: 402 3rd Qu.: 277 3rd Qu.: 190 3rd Qu.: 230
## Max. :123758 Max. :202629 Max. :270292 Max. :144030
## HPGL0267 HPGL0268 HPGL0269 HPGL0270
## Min. : 0 Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 10 1st Qu.: 17 1st Qu.: 9 1st Qu.: 4
## Median : 68 Median : 78 Median : 41 Median : 16
## Mean : 2672 Mean : 3353 Mean : 2779 Mean : 5050
## 3rd Qu.: 510 3rd Qu.: 586 3rd Qu.: 279 3rd Qu.: 79
## Max. :1469969 Max. :1859203 Max. :2378767 Max. :5953392
## HPGL0271 HPGL0272 HPGL0273 HPGL0274
## Min. : 0 Min. : 0 Min. : 0 Min. : 0
## 1st Qu.: 1 1st Qu.: 30 1st Qu.: 3 1st Qu.: 6
## Median : 10 Median : 210 Median : 15 Median : 21
## Mean : 149 Mean : 2586 Mean : 1815 Mean : 1977
## 3rd Qu.: 165 3rd Qu.: 912 3rd Qu.: 135 3rd Qu.: 66
## Max. :15922 Max. :1198103 Max. :1373655 Max. :1886309head(fData(experiment))## HPGL0259 HPGL0260 HPGL0261 HPGL0262 HPGL0263 HPGL0264 HPGL0265
## gene0 11 217 79 6 230 135 359
## gene1 54 141 115 46 20 23 13
## gene10 12 80 33 4 22 36 53
## gene100 0 9 2 1 3 4 1
## gene1000 892 205 33 14 261 366 1830
## gene1001 409 213 74 3 167 32 19
## HPGL0266 HPGL0267 HPGL0268 HPGL0269 HPGL0270 HPGL0271 HPGL0272
## gene0 242 19 55 38 29 1 70
## gene1 30 18 48 25 5 0 35
## gene10 61 25 55 27 16 1 21
## gene100 5 1 2 1 1 0 17
## gene1000 2205 6 29 11 8 11 217
## gene1001 19 96 3 2 1 0 5
## HPGL0273 HPGL0274
## gene0 15 24
## gene1 3 5
## gene10 17 16
## gene100 9 5
## gene1000 68 35
## gene1001 0 7head(pData(experiment))## sample condition batch strain color
## HPGL0259 HPGL0259 t0 9 5448 yellow4
## HPGL0260 HPGL0260 t1 9 5448 darkgreen
## HPGL0261 HPGL0261 t2 9 5448 red
## HPGL0262 HPGL0262 t3 9 5448 darkblue
## HPGL0263 HPGL0263 t0 11 5448 yellow4
## HPGL0264 HPGL0264 t1 11 5448 darkgreen
## counts
## HPGL0259 data/count_tables/5448-l1t0.count.gz
## HPGL0260 data/count_tables/5448-l1t1.count.gz
## HPGL0261 data/count_tables/5448-l1t2.count.gz
## HPGL0262 data/count_tables/5448-l1t3.count.gz
## HPGL0263 data/count_tables/5448-l2t0.count.gz
## HPGL0264 data/count_tables/5448-l2t1.count.gzGraph metrics
I can use my normalization toy on this data, but that seems like a terrible idea
all_expt = expt_subset(experiment, "")
metrics = graph_metrics(all_expt, out_type="cpm", norm_type="quant", filter="log2")## [1] "Filtering low counts"
## [1] "Low count filtering cost: 79 gene(s)."
## [1] "Applying normalization: quant"
## [1] "Setting output type as: cpm"
## [1] "Applying: log2 filter"
## [1] "Graphing number of non-zero genes with respect to CPM by library."
## [1] "Graphing library sizes."
## [1] "Adding log10"
## [1] "Graphing a raw data boxplot on log scale."## Using id as id variables## [1] "Graphing a normalized boxplot."## Using id as id variables## [1] "Graphing a raw-data correlation heatmap."
## [1] "Graphing a raw-data standard median correlation."
## [1] "Graphing a normalized correlation heatmap."
## [1] "Graphing a normalized standard median correlation."
## [1] "Graphing a raw-data distance heatmap."
## [1] "Generating distance matrix using: euclidean"
## [1] "Graphing a raw-data standard median distance."
## [1] "Graphing a normalized distance heatmap."
## [1] "Generating distance matrix using: euclidean"
## [1] "Graphing a normalized standard median distance."
## [1] "Graphing a PCA plot of the raw data."
## propVar cumPropVar cond.R2 batch.R2
## 1 75.22 75.22 17.91 58.41
## 2 13.72 88.94 14.72 58.18
## 3 7.59 96.53 7.78 63.84
## 4 2.46 98.99 30.92 33.18
## 5 0.40 99.39 31.63 45.60
## 6 0.23 99.62 23.14 1.25
## 7 0.13 99.75 11.37 6.12
## 8 0.08 99.83 41.10 0.64
## 9 0.07 99.90 6.87 1.30
## 10 0.04 99.94 22.56 2.24
## 11 0.03 99.97 20.44 0.47
## 12 0.02 99.99 26.94 0.41
## 13 0.01 100.00 20.94 0.01
## 14 0.01 100.01 21.88 10.57
## 15 0.00 100.01 1.81 17.77
## [1] "Graphing a PCA plot of the normalized data."
## propVar cumPropVar cond.R2 batch.R2
## 1 29.69 29.69 0.10 99.68
## 2 24.07 53.76 0.15 99.63
## 3 15.00 68.76 1.12 96.67
## 4 7.57 76.33 56.38 0.33
## 5 6.22 82.55 38.02 1.27
## 6 3.35 85.90 22.66 1.77
## 7 2.84 88.74 21.80 0.23
## 8 2.49 91.23 37.76 0.20
## 9 2.08 93.31 7.55 0.15
## 10 1.67 94.98 14.03 0.01
## 11 1.47 96.45 7.79 0.01
## 12 1.13 97.58 23.58 0.03
## 13 0.92 98.50 40.40 0.03
## 14 0.90 99.40 15.94 0.01
## 15 0.60 100.00 12.72 0.00metrics$norm_pcaplot
head(exprs(all_expt$expressionset))## HPGL0259 HPGL0260 HPGL0261 HPGL0262 HPGL0263 HPGL0264 HPGL0265
## gene0 11 217 79 6 230 135 359
## gene1 54 141 115 46 20 23 13
## gene10 12 80 33 4 22 36 53
## gene100 0 9 2 1 3 4 1
## gene1000 892 205 33 14 261 366 1830
## gene1001 409 213 74 3 167 32 19
## HPGL0266 HPGL0267 HPGL0268 HPGL0269 HPGL0270 HPGL0271 HPGL0272
## gene0 242 19 55 38 29 1 70
## gene1 30 18 48 25 5 0 35
## gene10 61 25 55 27 16 1 21
## gene100 5 1 2 1 1 0 17
## gene1000 2205 6 29 11 8 11 217
## gene1001 19 96 3 2 1 0 5
## HPGL0273 HPGL0274
## gene0 15 24
## gene1 3 5
## gene10 17 16
## gene100 9 5
## gene1000 68 35
## gene1001 0 7head(gene_annotations)## start end width strand Name protein_id ID
## M5005_Spy_0001 202 1557 1356 + dnaA <NA> gene0
## M5005_Spy_0002 1712 2848 1137 + dnaN <NA> gene1
## M5005_Spy_0003 2923 3120 198 + M5005_Spy_0003 <NA> gene2
## M5005_Spy_0004 3450 4565 1116 + ychF <NA> gene3
## M5005_Spy_0005 4635 5204 570 + pth <NA> gene4
## M5005_Spy_0006 5207 8710 3504 + trcF <NA> gene5id_gene_annotations = gene_annotations
rownames(id_gene_annotations) = id_gene_annotations$ID
all_qrpkm = my_norm(expt=all_expt, norm="quant", filter="log2", filter_low=FALSE, out_type="rpkm", annotations=id_gene_annotations)$counts## [1] "Applying normalization: quant"
## [1] "Setting output type as: rpkm"
## [1] "Applying: log2 filter"all_norm_expt = all_expt
## I did that so that I can replace the expressionset with the
## normalized data more easily...
normalized_expressionset = all_norm_expt$expressionset
exprs(normalized_expressionset) = all_qrpkm
all_norm_expt$expressionset = normalized_expressionset
all_design = data.frame(all_norm_expt$design)
all_design$longnames = paste(all_design$strain, all_design$batch, all_design$condition, sep="_")
my_boxplot(expt=all_norm_expt, names=all_design$longnames)## Using id as id variables
my_disheat(expt=all_norm_expt, names=all_design$longnames)## [1] "Generating distance matrix using: euclidean"
my_corheat(expt=all_norm_expt, names=all_design$longnames)
Get comparisons for Yoann
## Compare library 1, t0 t1
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='9' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t0_vs_lib1t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 76.04, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8532 0.8757
## sample estimates:
## cor
## 0.8649
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.7428 -0.7511 0.0687 0.7315 5.2811
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.4408 0.0641 6.88 8.2e-12 ***
## first 0.9347 0.0108 86.86 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.19
## Multiple R-squared: 0.796, Adjusted R-squared: 0.796
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1386 weights are ~= 1. The remaining 564 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0125 0.7440 0.9090 0.8150 0.9750 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.00
## 1st Qu.: 3.802 1st Qu.: 3.38
## Median : 5.515 Median : 5.72
## Mean : 5.434 Mean : 5.43
## 3rd Qu.: 7.073 3rd Qu.: 7.20
## Max. :17.933 Max. :17.93
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179329570707796 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 76.04, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8532 0.8757
## sample estimates:
## cor
## 0.8649dev.off()## Cairo
## 2## Compare library 1, t0 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='9' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t0_vs_lib1t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 55.28, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7635 0.7982
## sample estimates:
## cor
## 0.7814
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -8.986 -1.016 0.036 1.032 9.713
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.8408 0.0857 9.81 <2e-16 ***
## first 0.8470 0.0144 58.96 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.57
## Multiple R-squared: 0.648, Adjusted R-squared: 0.648
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1381 weights are ~= 1. The remaining 569 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0213 0.7950 0.9150 0.8460 0.9780 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.114
## 1st Qu.: 3.802 1st Qu.: 3.283
## Median : 5.515 Median : 5.669
## Mean : 5.434 Mean : 5.431
## 3rd Qu.: 7.073 3rd Qu.: 7.236
## Max. :17.933 Max. :17.933
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178269487069305 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 55.28, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7635 0.7982
## sample estimates:
## cor
## 0.7814dev.off()## Cairo
## 2## Compare library 1, t0 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='9' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t0_vs_lib1t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 33.07, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.5704 0.6273
## sample estimates:
## cor
## 0.5996
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -8.1103 -1.3636 0.0909 1.3546 10.6580
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.0262 0.1115 18.2 <2e-16 ***
## first 0.6249 0.0187 33.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.03
## Multiple R-squared: 0.374, Adjusted R-squared: 0.374
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1362 weights are ~= 1. The remaining 588 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0602 0.8100 0.9270 0.8670 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.465
## 1st Qu.: 3.802 1st Qu.: 3.419
## Median : 5.515 Median : 5.195
## Mean : 5.434 Mean : 5.431
## 3rd Qu.: 7.073 3rd Qu.: 7.256
## Max. :17.933 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178269445137466 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 33.07, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.5704 0.6273
## sample estimates:
## cor
## 0.5996dev.off()## Cairo
## 2## Compare library 1, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='9' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t1_vs_lib1t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 117.7, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.9306 0.9416
## sample estimates:
## cor
## 0.9363
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -3.769 -0.521 0.013 0.538 4.698
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.28221 0.04485 6.29 3.8e-10 ***
## first 0.94412 0.00745 126.65 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.857
## Multiple R-squared: 0.893, Adjusted R-squared: 0.893
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1397 weights are ~= 1. The remaining 553 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0477 0.7470 0.9030 0.8190 0.9720 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.114
## 1st Qu.: 3.38 1st Qu.: 3.283
## Median : 5.72 Median : 5.669
## Mean : 5.43 Mean : 5.431
## 3rd Qu.: 7.20 3rd Qu.: 7.236
## Max. :17.93 Max. :17.933
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179329570707796 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 117.7, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.9306 0.9416
## sample estimates:
## cor
## 0.9363dev.off()## Cairo
## 2## Compare library 1, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='9' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t1_vs_lib1t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 117.7, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.9306 0.9416
## sample estimates:
## cor
## 0.9363
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -3.769 -0.521 0.013 0.538 4.698
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.28221 0.04485 6.29 3.8e-10 ***
## first 0.94412 0.00745 126.65 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.857
## Multiple R-squared: 0.893, Adjusted R-squared: 0.893
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1397 weights are ~= 1. The remaining 553 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0477 0.7470 0.9030 0.8190 0.9720 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.114
## 1st Qu.: 3.38 1st Qu.: 3.283
## Median : 5.72 Median : 5.669
## Mean : 5.43 Mean : 5.431
## 3rd Qu.: 7.20 3rd Qu.: 7.236
## Max. :17.93 Max. :17.933
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179329570707796 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 117.7, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.9306 0.9416
## sample estimates:
## cor
## 0.9363dev.off()## Cairo
## 2## Compare library 1, t1 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='9' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t1_vs_lib1t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 38.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6362 0.6861
## sample estimates:
## cor
## 0.6619
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.1056 -1.1553 0.0727 1.2184 7.5861
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.8635 0.0994 18.8 <2e-16 ***
## first 0.6618 0.0166 40.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.88
## Multiple R-squared: 0.461, Adjusted R-squared: 0.461
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1371 weights are ~= 1. The remaining 579 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.167 0.789 0.913 0.849 0.975 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.465
## 1st Qu.: 3.38 1st Qu.: 3.419
## Median : 5.72 Median : 5.195
## Mean : 5.43 Mean : 5.431
## 3rd Qu.: 7.20 3rd Qu.: 7.256
## Max. :17.93 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.99
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.99
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179329570707796 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 38.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6362 0.6861
## sample estimates:
## cor
## 0.6619dev.off()## Cairo
## 2## Compare library 1, t2 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t2')|(batch=='9' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib1t2_vs_lib1t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 44.57, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6879 0.7319
## sample estimates:
## cor
## 0.7106
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.2890 -1.0666 0.0814 1.0436 7.0389
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.6600 0.0902 18.4 <2e-16 ***
## first 0.7055 0.0150 47.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.74
## Multiple R-squared: 0.539, Adjusted R-squared: 0.539
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1377 weights are ~= 1. The remaining 573 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.167 0.765 0.915 0.837 0.978 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.114 Min. : 0.465
## 1st Qu.: 3.283 1st Qu.: 3.419
## Median : 5.669 Median : 5.195
## Mean : 5.431 Mean : 5.431
## 3rd Qu.: 7.236 3rd Qu.: 7.256
## Max. :17.933 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178188286158731 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 44.57, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6879 0.7319
## sample estimates:
## cor
## 0.7106dev.off()## Cairo
## 2## Compare library 2, t0 t1
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='11' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t0_vs_lib2t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 94.9, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8985 0.9143
## sample estimates:
## cor
## 0.9067
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.5153 -0.5826 0.0463 0.5496 5.0923
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.54678 0.05232 10.4 <2e-16 ***
## first 0.90656 0.00876 103.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.962
## Multiple R-squared: 0.848, Adjusted R-squared: 0.848
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1381 weights are ~= 1. The remaining 569 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0365 0.7150 0.9040 0.8100 0.9770 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.379
## 1st Qu.: 3.588 1st Qu.: 3.547
## Median : 5.016 Median : 5.061
## Mean : 5.400 Mean : 5.403
## 3rd Qu.: 7.042 3rd Qu.: 7.065
## Max. :18.567 Max. :18.836
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.96
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.96
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018568737769307 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 94.9, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8985 0.9143
## sample estimates:
## cor
## 0.9067dev.off()## Cairo
## 2## Compare library 2, t0 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='11' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t0_vs_lib2t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 78.1, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8594 0.8809
## sample estimates:
## cor
## 0.8706
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.8738 -0.6679 0.0821 0.6352 4.7825
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.69887 0.05921 11.8 <2e-16 ***
## first 0.88744 0.00989 89.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.1
## Multiple R-squared: 0.806, Adjusted R-squared: 0.805
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1398 weights are ~= 1. The remaining 552 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0199 0.6670 0.8880 0.7940 0.9790 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.394
## 1st Qu.: 3.588 1st Qu.: 3.619
## Median : 5.016 Median : 5.061
## Mean : 5.400 Mean : 5.403
## 3rd Qu.: 7.042 3rd Qu.: 7.033
## Max. :18.567 Max. :18.161
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.96
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.96
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00183004737479312 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 78.1, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8594 0.8809
## sample estimates:
## cor
## 0.8706dev.off()## Cairo
## 2## Compare library 2, t0 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='11' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t0_vs_lib2t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 78.92, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8618 0.8830
## sample estimates:
## cor
## 0.8728
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.6731 -0.6375 0.0311 0.6693 6.0657
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.71460 0.05929 12.1 <2e-16 ***
## first 0.87939 0.00997 88.2 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.1
## Multiple R-squared: 0.803, Adjusted R-squared: 0.802
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1370 weights are ~= 1. The remaining 580 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0247 0.7510 0.9080 0.8180 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.228
## 1st Qu.: 3.588 1st Qu.: 3.574
## Median : 5.016 Median : 4.986
## Mean : 5.400 Mean : 5.397
## 3rd Qu.: 7.042 3rd Qu.: 6.988
## Max. :18.567 Max. :17.837
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.98
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.98
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00183392150499062 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 78.92, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8618 0.8830
## sample estimates:
## cor
## 0.8728dev.off()## Cairo
## 2## Compare library 2, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t1')|(batch=='11' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t1_vs_lib2t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 81.15, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8679 0.8882
## sample estimates:
## cor
## 0.8785
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.19184 -0.59324 -0.00627 0.65116 6.39578
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.6669 0.0562 11.9 <2e-16 ***
## first 0.8878 0.0094 94.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.05
## Multiple R-squared: 0.82, Adjusted R-squared: 0.82
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1400 weights are ~= 1. The remaining 550 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0069 0.6730 0.8970 0.7920 0.9730 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.379 Min. : 0.394
## 1st Qu.: 3.547 1st Qu.: 3.619
## Median : 5.061 Median : 5.061
## Mean : 5.403 Mean : 5.403
## 3rd Qu.: 7.065 3rd Qu.: 7.033
## Max. :18.836 Max. :18.161
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184565259315678 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 81.15, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8679 0.8882
## sample estimates:
## cor
## 0.8785dev.off()## Cairo
## 2## Compare library 2, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t1')|(batch=='11' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t1_vs_lib2t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 81.15, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8679 0.8882
## sample estimates:
## cor
## 0.8785
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.19184 -0.59324 -0.00627 0.65116 6.39578
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.6669 0.0562 11.9 <2e-16 ***
## first 0.8878 0.0094 94.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.05
## Multiple R-squared: 0.82, Adjusted R-squared: 0.82
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1400 weights are ~= 1. The remaining 550 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0069 0.6730 0.8970 0.7920 0.9730 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.379 Min. : 0.394
## 1st Qu.: 3.547 1st Qu.: 3.619
## Median : 5.061 Median : 5.061
## Mean : 5.403 Mean : 5.403
## 3rd Qu.: 7.065 3rd Qu.: 7.033
## Max. :18.836 Max. :18.161
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 1
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184565259315678 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 81.15, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8679 0.8882
## sample estimates:
## cor
## 0.8785dev.off()## Cairo
## 2## Compare library 2, t1 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t1')|(batch=='11' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t1_vs_lib2t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 86.04, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8801 0.8987
## sample estimates:
## cor
## 0.8898
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.2562 -0.6154 0.0372 0.5886 7.1339
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.64001 0.05389 11.9 <2e-16 ***
## first 0.88706 0.00904 98.1 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1
## Multiple R-squared: 0.831, Adjusted R-squared: 0.831
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1419 weights are ~= 1. The remaining 531 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0034 0.6820 0.8700 0.7830 0.9720 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.379 Min. : 0.228
## 1st Qu.: 3.547 1st Qu.: 3.574
## Median : 5.061 Median : 4.986
## Mean : 5.403 Mean : 5.397
## 3rd Qu.: 7.065 3rd Qu.: 6.988
## Max. :18.836 Max. :17.837
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018607479071282 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 86.04, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8801 0.8987
## sample estimates:
## cor
## 0.8898dev.off()## Cairo
## 2## Compare library 2, t2 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t2')|(batch=='11' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib2t2_vs_lib2t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 86.41, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8810 0.8994
## sample estimates:
## cor
## 0.8906
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.8073 -0.5463 -0.0191 0.5755 7.2175
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.5037 0.0529 9.52 <2e-16 ***
## first 0.9087 0.0088 103.28 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.973
## Multiple R-squared: 0.846, Adjusted R-squared: 0.846
## Convergence in 9 IRWLS iterations
##
## Robustness weights:
## 1394 weights are ~= 1. The remaining 556 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0012 0.6590 0.8950 0.7890 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.394 Min. : 0.228
## 1st Qu.: 3.619 1st Qu.: 3.574
## Median : 5.061 Median : 4.986
## Mean : 5.403 Mean : 5.397
## 3rd Qu.: 7.033 3rd Qu.: 6.988
## Max. :18.161 Max. :17.837
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179328750500349 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 86.41, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8810 0.8994
## sample estimates:
## cor
## 0.8906dev.off()## Cairo
## 2## Compare library 3, t0 t1
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t0')|(batch=='12' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t0_vs_lib3t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 94.2, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8972 0.9132
## sample estimates:
## cor
## 0.9055
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.1828 -0.6195 0.0638 0.5913 5.4309
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.36308 0.05151 7.05 2.5e-12 ***
## first 0.93731 0.00852 110.02 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.989
## Multiple R-squared: 0.861, Adjusted R-squared: 0.861
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1417 weights are ~= 1. The remaining 533 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.019 0.662 0.889 0.789 0.972 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.323 Min. : 0.077
## 1st Qu.: 3.359 1st Qu.: 3.414
## Median : 5.292 Median : 5.202
## Mean : 5.408 Mean : 5.400
## 3rd Qu.: 7.054 3rd Qu.: 7.053
## Max. :18.768 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186907666189841 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 94.2, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8972 0.9132
## sample estimates:
## cor
## 0.9055dev.off()## Cairo
## 2## Compare library 3, t0 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t0')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t0_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 71.82, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8393 0.8637
## sample estimates:
## cor
## 0.852
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -8.5766 -0.7657 0.0479 0.8033 6.0277
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.6697 0.0644 10.4 <2e-16 ***
## first 0.8857 0.0107 82.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.24
## Multiple R-squared: 0.778, Adjusted R-squared: 0.778
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1393 weights are ~= 1. The remaining 557 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0056 0.7270 0.9100 0.8140 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.323 Min. : 0.203
## 1st Qu.: 3.359 1st Qu.: 3.411
## Median : 5.292 Median : 5.158
## Mean : 5.408 Mean : 5.407
## 3rd Qu.: 7.054 3rd Qu.: 7.182
## Max. :18.768 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.99
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.99
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185649629004134 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 71.82, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8393 0.8637
## sample estimates:
## cor
## 0.852dev.off()## Cairo
## 2## Compare library 3, t0 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t0')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t0_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 59.79, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7883 0.8196
## sample estimates:
## cor
## 0.8045
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.6307 -0.8807 0.0494 0.8632 5.8009
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.0804 0.0732 14.8 <2e-16 ***
## first 0.8159 0.0122 66.8 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.41
## Multiple R-squared: 0.698, Adjusted R-squared: 0.698
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1395 weights are ~= 1. The remaining 555 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0981 0.7230 0.9080 0.8130 0.9740 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.323 Min. : 0.351
## 1st Qu.: 3.359 1st Qu.: 3.549
## Median : 5.292 Median : 5.274
## Mean : 5.408 Mean : 5.425
## 3rd Qu.: 7.054 3rd Qu.: 7.015
## Max. :18.768 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.84
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.84
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184450670225201 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 59.79, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7883 0.8196
## sample estimates:
## cor
## 0.8045dev.off()## Cairo
## 2## Compare library 3, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t1')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t1_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 93.38, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8957 0.9119
## sample estimates:
## cor
## 0.9041
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.9503 -0.6278 0.0289 0.5718 5.1039
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.48551 0.05151 9.43 <2e-16 ***
## first 0.92151 0.00856 107.63 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.989
## Multiple R-squared: 0.857, Adjusted R-squared: 0.857
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1402 weights are ~= 1. The remaining 548 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0045 0.7120 0.8970 0.7990 0.9700 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.077 Min. : 0.203
## 1st Qu.: 3.414 1st Qu.: 3.411
## Median : 5.202 Median : 5.158
## Mean : 5.400 Mean : 5.407
## 3rd Qu.: 7.053 3rd Qu.: 7.182
## Max. :18.094 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.93
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.93
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186907779017891 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 93.38, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8957 0.9119
## sample estimates:
## cor
## 0.9041dev.off()## Cairo
## 2## Compare library 3, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t1')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t1_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 93.38, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8957 0.9119
## sample estimates:
## cor
## 0.9041
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.9503 -0.6278 0.0289 0.5718 5.1039
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.48551 0.05151 9.43 <2e-16 ***
## first 0.92151 0.00856 107.63 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 0.989
## Multiple R-squared: 0.857, Adjusted R-squared: 0.857
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1402 weights are ~= 1. The remaining 548 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0045 0.7120 0.8970 0.7990 0.9700 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.077 Min. : 0.203
## 1st Qu.: 3.414 1st Qu.: 3.411
## Median : 5.202 Median : 5.158
## Mean : 5.400 Mean : 5.407
## 3rd Qu.: 7.053 3rd Qu.: 7.182
## Max. :18.094 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.93
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.93
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186907779017891 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 93.38, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8957 0.9119
## sample estimates:
## cor
## 0.9041dev.off()## Cairo
## 2## Compare library 3, t1 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t1')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t1_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 66.34, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8184 0.8457
## sample estimates:
## cor
## 0.8326
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.5047 -0.7908 0.0864 0.7869 4.9181
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.9896 0.0678 14.6 <2e-16 ***
## first 0.8345 0.0113 73.8 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.31
## Multiple R-squared: 0.739, Adjusted R-squared: 0.738
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1402 weights are ~= 1. The remaining 548 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0774 0.7140 0.8990 0.8060 0.9720 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.077 Min. : 0.351
## 1st Qu.: 3.414 1st Qu.: 3.549
## Median : 5.202 Median : 5.274
## Mean : 5.400 Mean : 5.425
## 3rd Qu.: 7.053 3rd Qu.: 7.015
## Max. :18.094 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00180161819658611 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 66.34, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8184 0.8457
## sample estimates:
## cor
## 0.8326dev.off()## Cairo
## 2## Compare library 3, t2 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t2')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib3t2_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 68.79, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8282 0.8541
## sample estimates:
## cor
## 0.8416
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.97178 -0.78853 -0.00347 0.77803 5.78562
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.9169 0.0678 13.5 <2e-16 ***
## first 0.8426 0.0113 74.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.3
## Multiple R-squared: 0.746, Adjusted R-squared: 0.746
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1398 weights are ~= 1. The remaining 552 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.107 0.753 0.900 0.822 0.972 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.203 Min. : 0.351
## 1st Qu.: 3.411 1st Qu.: 3.549
## Median : 5.158 Median : 5.274
## Mean : 5.407 Mean : 5.425
## 3rd Qu.: 7.182 3rd Qu.: 7.015
## Max. :18.768 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.83
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.83
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185649629004134 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 68.79, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.8282 0.8541
## sample estimates:
## cor
## 0.8416dev.off()## Cairo
## 2## Compare library 4, t0 t1
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t0')|(batch=='34' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t0_vs_lib4t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 31.16, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.5464 0.6056
## sample estimates:
## cor
## 0.5768
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.84528 -1.55232 0.00655 1.49988 9.69377
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.9174 0.1221 15.7 <2e-16 ***
## first 0.6290 0.0202 31.1 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.342, Adjusted R-squared: 0.341
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1367 weights are ~= 1. The remaining 583 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.123 0.815 0.938 0.879 0.979 0.998
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.696 Min. : 0.175
## 1st Qu.: 3.543 1st Qu.: 3.228
## Median : 5.193 Median : 5.470
## Mean : 5.488 Mean : 5.411
## 3rd Qu.: 7.220 3rd Qu.: 7.144
## Max. :17.984 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.35
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.35
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178091210866743 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 31.16, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.5464 0.6056
## sample estimates:
## cor
## 0.5768dev.off()## Cairo
## 2## Compare library 4, t0 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t0')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t0_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.32, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4786 0.5442
## sample estimates:
## cor
## 0.5121
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.1240 -1.5239 -0.0143 1.4621 8.4972
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.3455 0.1233 19 <2e-16 ***
## first 0.5539 0.0205 27 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.22
## Multiple R-squared: 0.28, Adjusted R-squared: 0.28
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1349 weights are ~= 1. The remaining 601 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.196 0.831 0.926 0.874 0.986 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.696 Min. : 0.426
## 1st Qu.: 3.543 1st Qu.: 3.470
## Median : 5.193 Median : 5.047
## Mean : 5.488 Mean : 5.426
## 3rd Qu.: 7.220 3rd Qu.: 7.196
## Max. :17.984 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.45
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.45
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00175582232469477 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.32, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4786 0.5442
## sample estimates:
## cor
## 0.5121dev.off()## Cairo
## 2## Compare library 4, t0 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t0')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t0_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.71, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4844 0.5495
## sample estimates:
## cor
## 0.5177
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.0024 -1.3975 0.0212 1.4615 10.1736
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4193 0.1193 20.3 <2e-16 ***
## first 0.5355 0.0199 27.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.15
## Multiple R-squared: 0.281, Adjusted R-squared: 0.28
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1341 weights are ~= 1. The remaining 609 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0947 0.8210 0.9470 0.8720 0.9820 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.696 Min. : 0.317
## 1st Qu.: 3.543 1st Qu.: 3.509
## Median : 5.193 Median : 5.211
## Mean : 5.488 Mean : 5.413
## 3rd Qu.: 7.220 3rd Qu.: 7.140
## Max. :17.984 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.35
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.35
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00176673798527041 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.71, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4844 0.5495
## sample estimates:
## cor
## 0.5177dev.off()## Cairo
## 2## Compare library 4, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t1')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t1_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 45.14, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6926 0.7361
## sample estimates:
## cor
## 0.715
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.569 -1.167 0.135 1.147 5.104
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.6447 0.0910 18.1 <2e-16 ***
## first 0.7176 0.0152 47.2 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.77
## Multiple R-squared: 0.541, Adjusted R-squared: 0.541
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1388 weights are ~= 1. The remaining 562 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.214 0.775 0.925 0.846 0.978 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.175 Min. : 0.426
## 1st Qu.: 3.228 1st Qu.: 3.470
## Median : 5.470 Median : 5.047
## Mean : 5.411 Mean : 5.426
## 3rd Qu.: 7.144 3rd Qu.: 7.196
## Max. :17.984 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.86
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.86
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178089400325604 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 45.14, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6926 0.7361
## sample estimates:
## cor
## 0.715dev.off()## Cairo
## 2## Compare library 4, t1 t2
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t1')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t1_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 45.14, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6926 0.7361
## sample estimates:
## cor
## 0.715
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.569 -1.167 0.135 1.147 5.104
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.6447 0.0910 18.1 <2e-16 ***
## first 0.7176 0.0152 47.2 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.77
## Multiple R-squared: 0.541, Adjusted R-squared: 0.541
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1388 weights are ~= 1. The remaining 562 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.214 0.775 0.925 0.846 0.978 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.175 Min. : 0.426
## 1st Qu.: 3.228 1st Qu.: 3.470
## Median : 5.470 Median : 5.047
## Mean : 5.411 Mean : 5.426
## 3rd Qu.: 7.144 3rd Qu.: 7.196
## Max. :17.984 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.86
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.86
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178089400325604 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 45.14, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6926 0.7361
## sample estimates:
## cor
## 0.715dev.off()## Cairo
## 2## Compare library 4, t1 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t1')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t1_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 51.23, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7380 0.7759
## sample estimates:
## cor
## 0.7576
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.5773 -0.9267 0.0492 1.0092 6.5616
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.4764 0.0808 18.3 <2e-16 ***
## first 0.7274 0.0135 54.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.57
## Multiple R-squared: 0.605, Adjusted R-squared: 0.605
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1386 weights are ~= 1. The remaining 564 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.151 0.747 0.913 0.830 0.979 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.175 Min. : 0.317
## 1st Qu.: 3.228 1st Qu.: 3.509
## Median : 5.470 Median : 5.211
## Mean : 5.411 Mean : 5.413
## 3rd Qu.: 7.144 3rd Qu.: 7.140
## Max. :17.984 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.98
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.98
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178089400325604 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 51.23, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7380 0.7759
## sample estimates:
## cor
## 0.7576dev.off()## Cairo
## 2## Compare library 4, t2 t3
comp = exprs(expt_subset(all_norm_expt, "(batch=='34' & condition=='t2')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/intertime/lib4t2_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 55.92, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7673 0.8014
## sample estimates:
## cor
## 0.785
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.0392 -1.0191 -0.0277 0.9553 6.7016
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.9782 0.0786 12.4 <2e-16 ***
## first 0.8122 0.0131 62.2 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.48
## Multiple R-squared: 0.668, Adjusted R-squared: 0.668
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1400 weights are ~= 1. The remaining 550 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.113 0.747 0.909 0.827 0.976 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.426 Min. : 0.317
## 1st Qu.: 3.470 1st Qu.: 3.509
## Median : 5.047 Median : 5.211
## Mean : 5.426 Mean : 5.413
## 3rd Qu.: 7.196 3rd Qu.: 7.140
## Max. :17.631 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.87
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.87
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00173143698094528 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 55.92, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.7673 0.8014
## sample estimates:
## cor
## 0.785dev.off()## Cairo
## 2Repeat correlations keeping time the same, but for each library set
## Compare t0, lib1 lib2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='11' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib2t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.3, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.398 0.470
## sample estimates:
## cor
## 0.4347
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.7738 -1.4167 -0.0312 1.3426 12.8503
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8673 0.1187 24.1 <2e-16 ***
## first 0.4451 0.0199 22.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.18
## Multiple R-squared: 0.211, Adjusted R-squared: 0.211
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1357 weights are ~= 1. The remaining 593 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0298 0.7950 0.9160 0.8530 0.9780 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.267
## 1st Qu.: 3.802 1st Qu.: 3.588
## Median : 5.515 Median : 5.016
## Mean : 5.434 Mean : 5.400
## 3rd Qu.: 7.073 3rd Qu.: 7.042
## Max. :17.933 Max. :18.567
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.67
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.67
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184613981388495 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.3, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.398 0.470
## sample estimates:
## cor
## 0.4347dev.off()## Cairo
## 2## Compare t0, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.695 -1.399 -0.019 1.339 11.619
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.506 0.119 21.0 <2e-16 ***
## first 0.513 0.020 25.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.26, Adjusted R-squared: 0.26
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1367 weights are ~= 1. The remaining 583 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0566 0.7770 0.9170 0.8460 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.323
## 1st Qu.: 3.802 1st Qu.: 3.359
## Median : 5.515 Median : 5.292
## Mean : 5.434 Mean : 5.408
## 3rd Qu.: 7.073 3rd Qu.: 7.054
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018662254030768 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921dev.off()## Cairo
## 2## Compare t0, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.695 -1.399 -0.019 1.339 11.619
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.506 0.119 21.0 <2e-16 ***
## first 0.513 0.020 25.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.26, Adjusted R-squared: 0.26
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1367 weights are ~= 1. The remaining 583 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0566 0.7770 0.9170 0.8460 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.323
## 1st Qu.: 3.802 1st Qu.: 3.359
## Median : 5.515 Median : 5.292
## Mean : 5.434 Mean : 5.408
## 3rd Qu.: 7.073 3rd Qu.: 7.054
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018662254030768 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921dev.off()## Cairo
## 2## Compare t0, lib1 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4094 0.4806
## sample estimates:
## cor
## 0.4457
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.1453 -1.5303 -0.0549 1.5114 9.4276
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9736 0.1224 24.3 <2e-16 ***
## first 0.4473 0.0205 21.8 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.23
## Multiple R-squared: 0.203, Adjusted R-squared: 0.203
## Convergence in 4 IRWLS iterations
##
## Robustness weights:
## 1347 weights are ~= 1. The remaining 603 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.144 0.832 0.930 0.874 0.981 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.696
## 1st Qu.: 3.802 1st Qu.: 3.543
## Median : 5.515 Median : 5.193
## Mean : 5.434 Mean : 5.488
## 3rd Qu.: 7.073 3rd Qu.: 7.220
## Max. :17.933 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.5
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.5
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178783602220492 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4094 0.4806
## sample estimates:
## cor
## 0.4457dev.off()## Cairo
## 2## Compare t0, lib2 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.03, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2820 0.3615
## sample estimates:
## cor
## 0.3223
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.907 -1.603 -0.127 1.624 13.454
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.2030 0.1286 24.9 <2e-16 ***
## first 0.3792 0.0216 17.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.37
## Multiple R-squared: 0.139, Adjusted R-squared: 0.139
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1395 weights are ~= 1. The remaining 555 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0388 0.7850 0.9280 0.8450 0.9790 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.323
## 1st Qu.: 3.588 1st Qu.: 3.359
## Median : 5.016 Median : 5.292
## Mean : 5.400 Mean : 5.408
## 3rd Qu.: 7.042 3rd Qu.: 7.054
## Max. :18.567 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185013296398497 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.03, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2820 0.3615
## sample estimates:
## cor
## 0.3223dev.off()## Cairo
## 2## Compare t0, lib2 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.72, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2559 0.3369
## sample estimates:
## cor
## 0.2969
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -4.958 -1.620 -0.263 1.735 12.205
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.7276 0.1297 28.7 <2e-16 ***
## first 0.3032 0.0217 13.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.38
## Multiple R-squared: 0.0945, Adjusted R-squared: 0.0941
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1335 weights are ~= 1. The remaining 615 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0678 0.8520 0.9370 0.8850 0.9820 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.696
## 1st Qu.: 3.588 1st Qu.: 3.543
## Median : 5.016 Median : 5.193
## Mean : 5.400 Mean : 5.488
## 3rd Qu.: 7.042 3rd Qu.: 7.220
## Max. :18.567 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.27
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.27
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00183004737479312 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.72, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2559 0.3369
## sample estimates:
## cor
## 0.2969dev.off()## Cairo
## 2## Compare t0, lib3 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib3t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.31, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2475 0.3289
## sample estimates:
## cor
## 0.2888
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.3849 -1.6005 -0.0988 1.6521 11.9325
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.8213 0.1271 30.1 <2e-16 ***
## first 0.2877 0.0212 13.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.39
## Multiple R-squared: 0.0897, Adjusted R-squared: 0.0893
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1346 weights are ~= 1. The remaining 604 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0759 0.8280 0.9340 0.8760 0.9810 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.323 Min. : 0.696
## 1st Qu.: 3.359 1st Qu.: 3.543
## Median : 5.292 Median : 5.193
## Mean : 5.408 Mean : 5.488
## 3rd Qu.: 7.054 3rd Qu.: 7.220
## Max. :18.768 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.33
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.33
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184450670225201 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.31, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2475 0.3289
## sample estimates:
## cor
## 0.2888dev.off()## Cairo
## 2## Compare t1, lib1 lib2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='11' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t1_vs_lib2t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.57, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4025 0.4742
## sample estimates:
## cor
## 0.4391
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.9144 -1.4779 0.0367 1.3578 13.0063
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.0403 0.1172 25.9 <2e-16 ***
## first 0.4211 0.0196 21.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.22
## Multiple R-squared: 0.2, Adjusted R-squared: 0.199
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1371 weights are ~= 1. The remaining 579 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0314 0.7940 0.9130 0.8570 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.379
## 1st Qu.: 3.38 1st Qu.: 3.547
## Median : 5.72 Median : 5.061
## Mean : 5.43 Mean : 5.403
## 3rd Qu.: 7.20 3rd Qu.: 7.065
## Max. :17.93 Max. :18.836
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.73
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.73
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00188356674764435 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.57, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4025 0.4742
## sample estimates:
## cor
## 0.4391dev.off()## Cairo
## 2## Compare t1, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='12' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t1_vs_lib3t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.54, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4820 0.5472
## sample estimates:
## cor
## 0.5153
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.9238 -1.3049 -0.0277 1.3119 11.1435
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.529 0.114 22.2 <2e-16 ***
## first 0.517 0.019 27.1 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.16
## Multiple R-squared: 0.283, Adjusted R-squared: 0.282
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1391 weights are ~= 1. The remaining 559 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0658 0.7570 0.8960 0.8350 0.9680 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.077
## 1st Qu.: 3.38 1st Qu.: 3.414
## Median : 5.72 Median : 5.202
## Mean : 5.43 Mean : 5.400
## 3rd Qu.: 7.20 3rd Qu.: 7.053
## Max. :17.93 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.71
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.71
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00180936633029751 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.54, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4820 0.5472
## sample estimates:
## cor
## 0.5153dev.off()## Cairo
## 2## Compare t1, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='12' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t1_vs_lib3t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.54, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4820 0.5472
## sample estimates:
## cor
## 0.5153
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.9238 -1.3049 -0.0277 1.3119 11.1435
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.529 0.114 22.2 <2e-16 ***
## first 0.517 0.019 27.1 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.16
## Multiple R-squared: 0.283, Adjusted R-squared: 0.282
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1391 weights are ~= 1. The remaining 559 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0658 0.7570 0.8960 0.8350 0.9680 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.077
## 1st Qu.: 3.38 1st Qu.: 3.414
## Median : 5.72 Median : 5.202
## Mean : 5.43 Mean : 5.400
## 3rd Qu.: 7.20 3rd Qu.: 7.053
## Max. :17.93 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.71
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.71
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00180936633029751 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.54, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4820 0.5472
## sample estimates:
## cor
## 0.5153dev.off()## Cairo
## 2## Compare t1, lib1 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t1')|(batch=='34' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t1_vs_lib4t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 41.76, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6631 0.7100
## sample estimates:
## cor
## 0.6873
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.3283 -1.0420 -0.0881 1.1454 10.7452
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.4982 0.0933 16.1 <2e-16 ***
## first 0.7057 0.0156 45.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 1.79
## Multiple R-squared: 0.519, Adjusted R-squared: 0.519
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1402 weights are ~= 1. The remaining 548 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0267 0.7260 0.9010 0.8130 0.9700 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.00 Min. : 0.175
## 1st Qu.: 3.38 1st Qu.: 3.228
## Median : 5.72 Median : 5.470
## Mean : 5.43 Mean : 5.411
## 3rd Qu.: 7.20 3rd Qu.: 7.144
## Max. :17.93 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.81
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.81
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00179841844841535 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 41.76, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.6631 0.7100
## sample estimates:
## cor
## 0.6873dev.off()## Cairo
## 2## Compare t1, lib2 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t1')|(batch=='12' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t1_vs_lib3t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.1, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3030 0.3813
## sample estimates:
## cor
## 0.3427
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.014 -1.530 -0.187 1.590 12.573
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.1293 0.1287 24.3 <2e-16 ***
## first 0.3946 0.0216 18.2 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.36
## Multiple R-squared: 0.149, Adjusted R-squared: 0.149
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1384 weights are ~= 1. The remaining 566 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0562 0.7960 0.9250 0.8480 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.379 Min. : 0.077
## 1st Qu.: 3.547 1st Qu.: 3.414
## Median : 5.061 Median : 5.202
## Mean : 5.403 Mean : 5.400
## 3rd Qu.: 7.065 3rd Qu.: 7.053
## Max. :18.836 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00187581747484414 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.1, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3030 0.3813
## sample estimates:
## cor
## 0.3427dev.off()## Cairo
## 2## Compare t1, lib2 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t1')|(batch=='34' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t1_vs_lib4t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.58, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3677 0.4419
## sample estimates:
## cor
## 0.4055
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.3491 -1.8168 -0.0519 1.6649 11.7601
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8493 0.1308 21.8 <2e-16 ***
## first 0.4509 0.0219 20.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.4
## Multiple R-squared: 0.183, Adjusted R-squared: 0.183
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1333 weights are ~= 1. The remaining 617 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0827 0.8630 0.9560 0.8860 0.9850 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.379 Min. : 0.175
## 1st Qu.: 3.547 1st Qu.: 3.228
## Median : 5.061 Median : 5.470
## Mean : 5.403 Mean : 5.411
## 3rd Qu.: 7.065 3rd Qu.: 7.144
## Max. :18.836 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186604230248504 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.58, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3677 0.4419
## sample estimates:
## cor
## 0.4055dev.off()## Cairo
## 2## Compare t1, lib3 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t1')|(batch=='34' & condition=='t1')")$expressionset)
pdf(file="figures/comparisons/interlib/lib3t1_vs_lib4t1_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 23.26, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4308 0.5003
## sample estimates:
## cor
## 0.4663
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.9716 -1.6050 -0.0715 1.5042 11.9105
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6173 0.1233 21.2 <2e-16 ***
## first 0.4964 0.0206 24.1 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.32
## Multiple R-squared: 0.237, Adjusted R-squared: 0.236
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1364 weights are ~= 1. The remaining 586 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.067 0.810 0.930 0.868 0.981 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.077 Min. : 0.175
## 1st Qu.: 3.414 1st Qu.: 3.228
## Median : 5.202 Median : 5.470
## Mean : 5.400 Mean : 5.411
## 3rd Qu.: 7.053 3rd Qu.: 7.144
## Max. :18.094 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.89
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.89
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018016170574973 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 23.26, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4308 0.5003
## sample estimates:
## cor
## 0.4663dev.off()## Cairo
## 2## Compare t2, lib1 lib2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t2')|(batch=='11' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t2_vs_lib2t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 18.69, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3516 0.4269
## sample estimates:
## cor
## 0.3899
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.3725 -1.4695 0.0437 1.4046 12.2327
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.3140 0.1182 28.0 <2e-16 ***
## first 0.3723 0.0197 18.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.26
## Multiple R-squared: 0.161, Adjusted R-squared: 0.16
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1364 weights are ~= 1. The remaining 586 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0513 0.7950 0.9220 0.8590 0.9770 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.114 Min. : 0.394
## 1st Qu.: 3.283 1st Qu.: 3.619
## Median : 5.669 Median : 5.061
## Mean : 5.431 Mean : 5.403
## 3rd Qu.: 7.236 3rd Qu.: 7.033
## Max. :17.933 Max. :18.161
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.73
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.73
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00180469380479209 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 18.69, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3516 0.4269
## sample estimates:
## cor
## 0.3899dev.off()## Cairo
## 2## Compare t2, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t2')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t2_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.81, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4859 0.5508
## sample estimates:
## cor
## 0.5191
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.3623 -1.3266 0.0111 1.2971 11.7596
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4693 0.1110 22.2 <2e-16 ***
## first 0.5307 0.0185 28.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.14
## Multiple R-squared: 0.303, Adjusted R-squared: 0.302
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1388 weights are ~= 1. The remaining 562 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0466 0.7580 0.9060 0.8300 0.9750 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.114 Min. : 0.203
## 1st Qu.: 3.283 1st Qu.: 3.411
## Median : 5.669 Median : 5.158
## Mean : 5.431 Mean : 5.407
## 3rd Qu.: 7.236 3rd Qu.: 7.182
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186541452225156 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.81, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4859 0.5508
## sample estimates:
## cor
## 0.5191dev.off()## Cairo
## 2## Compare t2, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t2')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t2_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.81, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4859 0.5508
## sample estimates:
## cor
## 0.5191
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.3623 -1.3266 0.0111 1.2971 11.7596
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4693 0.1110 22.2 <2e-16 ***
## first 0.5307 0.0185 28.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.14
## Multiple R-squared: 0.303, Adjusted R-squared: 0.302
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1388 weights are ~= 1. The remaining 562 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0466 0.7580 0.9060 0.8300 0.9750 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.114 Min. : 0.203
## 1st Qu.: 3.283 1st Qu.: 3.411
## Median : 5.669 Median : 5.158
## Mean : 5.431 Mean : 5.407
## 3rd Qu.: 7.236 3rd Qu.: 7.182
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.77
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00186541452225156 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 26.81, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4859 0.5508
## sample estimates:
## cor
## 0.5191dev.off()## Cairo
## 2## Compare t2, lib1 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t2')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t2_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.2, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4459 0.5142
## sample estimates:
## cor
## 0.4808
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -8.4024 -1.4273 -0.0421 1.3926 10.6812
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.7851 0.1178 23.6 <2e-16 ***
## first 0.4785 0.0197 24.3 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.24
## Multiple R-squared: 0.241, Adjusted R-squared: 0.241
## Convergence in 5 IRWLS iterations
##
## Robustness weights:
## 1375 weights are ~= 1. The remaining 575 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0923 0.7990 0.9040 0.8560 0.9740 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.114 Min. : 0.426
## 1st Qu.: 3.283 1st Qu.: 3.470
## Median : 5.669 Median : 5.047
## Mean : 5.431 Mean : 5.426
## 3rd Qu.: 7.236 3rd Qu.: 7.196
## Max. :17.933 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.95
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.95
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178188286158731 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.2, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4459 0.5142
## sample estimates:
## cor
## 0.4808dev.off()## Cairo
## 2## Compare t2, lib2 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t2')|(batch=='12' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t2_vs_lib3t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.18, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2850 0.3644
## sample estimates:
## cor
## 0.3253
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.31 -1.54 -0.20 1.61 13.68
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.1847 0.1307 24.4 <2e-16 ***
## first 0.3816 0.0219 17.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.39
## Multiple R-squared: 0.138, Adjusted R-squared: 0.138
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1361 weights are ~= 1. The remaining 589 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0364 0.8090 0.9290 0.8550 0.9810 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.394 Min. : 0.203
## 1st Qu.: 3.619 1st Qu.: 3.411
## Median : 5.061 Median : 5.158
## Mean : 5.403 Mean : 5.407
## 3rd Qu.: 7.033 3rd Qu.: 7.182
## Max. :18.161 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.97
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185649629004134 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.18, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2850 0.3644
## sample estimates:
## cor
## 0.3253dev.off()## Cairo
## 2## Compare t2, lib2 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t2')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t2_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 14.33, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2681 0.3484
## sample estimates:
## cor
## 0.3088
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.631 -1.544 -0.184 1.616 12.530
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.4357 0.1317 26.1 <2e-16 ***
## first 0.3415 0.0221 15.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.41
## Multiple R-squared: 0.113, Adjusted R-squared: 0.113
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1351 weights are ~= 1. The remaining 599 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0626 0.8170 0.9300 0.8670 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.394 Min. : 0.426
## 1st Qu.: 3.619 1st Qu.: 3.470
## Median : 5.061 Median : 5.047
## Mean : 5.403 Mean : 5.426
## 3rd Qu.: 7.033 3rd Qu.: 7.196
## Max. :18.161 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.78
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.78
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00177674991165709 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 14.33, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2681 0.3484
## sample estimates:
## cor
## 0.3088dev.off()## Cairo
## 2## Compare t2, lib3 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t2')|(batch=='34' & condition=='t2')")$expressionset)
pdf(file="figures/comparisons/interlib/lib3t2_vs_lib4t2_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.5, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3106 0.3885
## sample estimates:
## cor
## 0.3502
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.782 -1.542 -0.147 1.539 13.362
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.2868 0.1258 26.1 <2e-16 ***
## first 0.3738 0.0211 17.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.37
## Multiple R-squared: 0.144, Adjusted R-squared: 0.144
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1359 weights are ~= 1. The remaining 591 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0402 0.7960 0.9300 0.8580 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.203 Min. : 0.426
## 1st Qu.: 3.411 1st Qu.: 3.470
## Median : 5.158 Median : 5.047
## Mean : 5.407 Mean : 5.426
## 3rd Qu.: 7.182 3rd Qu.: 7.196
## Max. :18.768 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.82
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.82
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185649629004134 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.5, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3106 0.3885
## sample estimates:
## cor
## 0.3502dev.off()## Cairo
## 2## Compare t3, lib1 lib2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t3')|(batch=='11' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t3_vs_lib2t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.53, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3112 0.3891
## sample estimates:
## cor
## 0.3508
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.8554 -1.4767 -0.0621 1.4332 12.2567
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.3638 0.1239 27.1 <2e-16 ***
## first 0.3527 0.0207 17.0 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.29
## Multiple R-squared: 0.135, Adjusted R-squared: 0.135
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1359 weights are ~= 1. The remaining 591 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0545 0.8140 0.9220 0.8610 0.9730 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.465 Min. : 0.228
## 1st Qu.: 3.419 1st Qu.: 3.574
## Median : 5.195 Median : 4.986
## Mean : 5.431 Mean : 5.397
## 3rd Qu.: 7.256 3rd Qu.: 6.988
## Max. :17.631 Max. :17.837
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.68
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.68
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00176084353535542 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 16.53, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3112 0.3891
## sample estimates:
## cor
## 0.3508dev.off()## Cairo
## 2## Compare t3, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t3')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t3_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.91, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3737 0.4475
## sample estimates:
## cor
## 0.4113
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.2710 -1.5220 -0.0796 1.5536 10.0757
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9866 0.1234 24.2 <2e-16 ***
## first 0.4330 0.0207 20.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.29
## Multiple R-squared: 0.189, Adjusted R-squared: 0.189
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1371 weights are ~= 1. The remaining 579 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.125 0.801 0.929 0.856 0.978 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.465 Min. : 0.351
## 1st Qu.: 3.419 1st Qu.: 3.549
## Median : 5.195 Median : 5.274
## Mean : 5.431 Mean : 5.425
## 3rd Qu.: 7.256 3rd Qu.: 7.015
## Max. :17.631 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00177431371371418 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.91, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3737 0.4475
## sample estimates:
## cor
## 0.4113dev.off()## Cairo
## 2## Compare t3, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t3')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t3_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.91, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3737 0.4475
## sample estimates:
## cor
## 0.4113
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.2710 -1.5220 -0.0796 1.5536 10.0757
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9866 0.1234 24.2 <2e-16 ***
## first 0.4330 0.0207 20.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.29
## Multiple R-squared: 0.189, Adjusted R-squared: 0.189
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1371 weights are ~= 1. The remaining 579 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.125 0.801 0.929 0.856 0.978 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.465 Min. : 0.351
## 1st Qu.: 3.419 1st Qu.: 3.549
## Median : 5.195 Median : 5.274
## Mean : 5.431 Mean : 5.425
## 3rd Qu.: 7.256 3rd Qu.: 7.015
## Max. :17.631 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.94
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00177431371371418 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 19.91, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3737 0.4475
## sample estimates:
## cor
## 0.4113dev.off()## Cairo
## 2## Compare t3, lib1 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t3')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t3_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 23.49, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4345 0.5037
## sample estimates:
## cor
## 0.4698
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -7.093 -1.418 -0.108 1.463 10.813
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.6718 0.1175 22.7 <2e-16 ***
## first 0.4910 0.0197 24.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.18
## Multiple R-squared: 0.249, Adjusted R-squared: 0.248
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1375 weights are ~= 1. The remaining 575 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0777 0.8000 0.9310 0.8540 0.9780 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.465 Min. : 0.317
## 1st Qu.: 3.419 1st Qu.: 3.509
## Median : 5.195 Median : 5.211
## Mean : 5.431 Mean : 5.413
## 3rd Qu.: 7.256 3rd Qu.: 7.140
## Max. :17.631 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.83
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.83
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00173143698094528 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 23.49, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4345 0.5037
## sample estimates:
## cor
## 0.4698dev.off()## Cairo
## 2## Compare t3, lib2 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t3')|(batch=='12' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t3_vs_lib3t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.29, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2872 0.3664
## sample estimates:
## cor
## 0.3274
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.93 -1.55 -0.04 1.52 12.20
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.3164 0.1285 25.8 <2e-16 ***
## first 0.3666 0.0216 16.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.35
## Multiple R-squared: 0.132, Adjusted R-squared: 0.132
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1360 weights are ~= 1. The remaining 590 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0632 0.8040 0.9270 0.8550 0.9790 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.228 Min. : 0.351
## 1st Qu.: 3.574 1st Qu.: 3.549
## Median : 4.986 Median : 5.274
## Mean : 5.397 Mean : 5.425
## 3rd Qu.: 6.988 3rd Qu.: 7.015
## Max. :17.837 Max. :18.094
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.74
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.74
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178654862887017 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.29, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2872 0.3664
## sample estimates:
## cor
## 0.3274dev.off()## Cairo
## 2## Compare t3, lib2 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t3')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t3_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.02, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3931 0.4655
## sample estimates:
## cor
## 0.43
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.134 -1.444 -0.158 1.538 12.308
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8282 0.1186 23.8 <2e-16 ***
## first 0.4506 0.0199 22.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.213, Adjusted R-squared: 0.212
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1360 weights are ~= 1. The remaining 590 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0414 0.8100 0.9360 0.8560 0.9820 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.228 Min. : 0.317
## 1st Qu.: 3.574 1st Qu.: 3.509
## Median : 4.986 Median : 5.211
## Mean : 5.397 Mean : 5.413
## 3rd Qu.: 6.988 3rd Qu.: 7.140
## Max. :17.837 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.84
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.84
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00176084353535542 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.02, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3931 0.4655
## sample estimates:
## cor
## 0.43dev.off()## Cairo
## 2## Compare t3, lib3 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t3')|(batch=='34' & condition=='t3')")$expressionset)
pdf(file="figures/comparisons/interlib/lib3t3_vs_lib4t3_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 17.13, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3226 0.3998
## sample estimates:
## cor
## 0.3618
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.810 -1.579 -0.128 1.549 13.245
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.3110 0.1262 26.2 <2e-16 ***
## first 0.3685 0.0212 17.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.33
## Multiple R-squared: 0.14, Adjusted R-squared: 0.139
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1357 weights are ~= 1. The remaining 593 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0384 0.8250 0.9310 0.8690 0.9800 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.351 Min. : 0.317
## 1st Qu.: 3.549 1st Qu.: 3.509
## Median : 5.274 Median : 5.211
## Mean : 5.425 Mean : 5.413
## 3rd Qu.: 7.015 3rd Qu.: 7.140
## Max. :18.094 Max. :17.631
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.89
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.89
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00177766890083 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 17.13, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.3226 0.3998
## sample estimates:
## cor
## 0.3618dev.off()## Cairo
## 2## Compare t0, lib1 lib2
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='11' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib2t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.3, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.398 0.470
## sample estimates:
## cor
## 0.4347
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.7738 -1.4167 -0.0312 1.3426 12.8503
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.8673 0.1187 24.1 <2e-16 ***
## first 0.4451 0.0199 22.4 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.18
## Multiple R-squared: 0.211, Adjusted R-squared: 0.211
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1357 weights are ~= 1. The remaining 593 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0298 0.7950 0.9160 0.8530 0.9780 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.267
## 1st Qu.: 3.802 1st Qu.: 3.588
## Median : 5.515 Median : 5.016
## Mean : 5.434 Mean : 5.400
## 3rd Qu.: 7.073 3rd Qu.: 7.042
## Max. :17.933 Max. :18.567
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.67
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.67
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184613981388495 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.3, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.398 0.470
## sample estimates:
## cor
## 0.4347dev.off()## Cairo
## 2## Compare t0, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.695 -1.399 -0.019 1.339 11.619
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.506 0.119 21.0 <2e-16 ***
## first 0.513 0.020 25.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.26, Adjusted R-squared: 0.26
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1367 weights are ~= 1. The remaining 583 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0566 0.7770 0.9170 0.8460 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.323
## 1st Qu.: 3.802 1st Qu.: 3.359
## Median : 5.515 Median : 5.292
## Mean : 5.434 Mean : 5.408
## 3rd Qu.: 7.073 3rd Qu.: 7.054
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018662254030768 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921dev.off()## Cairo
## 2## Compare t0, lib1 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.695 -1.399 -0.019 1.339 11.619
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.506 0.119 21.0 <2e-16 ***
## first 0.513 0.020 25.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.19
## Multiple R-squared: 0.26, Adjusted R-squared: 0.26
## Convergence in 7 IRWLS iterations
##
## Robustness weights:
## 1367 weights are ~= 1. The remaining 583 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0566 0.7770 0.9170 0.8460 0.9760 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.323
## 1st Qu.: 3.802 1st Qu.: 3.359
## Median : 5.515 Median : 5.292
## Mean : 5.434 Mean : 5.408
## 3rd Qu.: 7.073 3rd Qu.: 7.054
## Max. :17.933 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.75
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.0018662254030768 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 24.95, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4577 0.5250
## sample estimates:
## cor
## 0.4921dev.off()## Cairo
## 2## Compare t0, lib1 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='9' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib1t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4094 0.4806
## sample estimates:
## cor
## 0.4457
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.1453 -1.5303 -0.0549 1.5114 9.4276
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.9736 0.1224 24.3 <2e-16 ***
## first 0.4473 0.0205 21.8 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.23
## Multiple R-squared: 0.203, Adjusted R-squared: 0.203
## Convergence in 4 IRWLS iterations
##
## Robustness weights:
## 1347 weights are ~= 1. The remaining 603 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.144 0.832 0.930 0.874 0.981 0.999
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.106 Min. : 0.696
## 1st Qu.: 3.802 1st Qu.: 3.543
## Median : 5.515 Median : 5.193
## Mean : 5.434 Mean : 5.488
## 3rd Qu.: 7.073 3rd Qu.: 7.220
## Max. :17.933 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.5
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.5
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00178783602220492 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 21.97, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4094 0.4806
## sample estimates:
## cor
## 0.4457dev.off()## Cairo
## 2## Compare t0, lib2 lib3
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='12' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t0_vs_lib3t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.03, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2820 0.3615
## sample estimates:
## cor
## 0.3223
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -6.907 -1.603 -0.127 1.624 13.454
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.2030 0.1286 24.9 <2e-16 ***
## first 0.3792 0.0216 17.6 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.37
## Multiple R-squared: 0.139, Adjusted R-squared: 0.139
## Convergence in 8 IRWLS iterations
##
## Robustness weights:
## 1395 weights are ~= 1. The remaining 555 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0388 0.7850 0.9280 0.8450 0.9790 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.323
## 1st Qu.: 3.588 1st Qu.: 3.359
## Median : 5.016 Median : 5.292
## Mean : 5.400 Mean : 5.408
## 3rd Qu.: 7.042 3rd Qu.: 7.054
## Max. :18.567 Max. :18.768
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.92
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00185013296398497 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 15.03, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2820 0.3615
## sample estimates:
## cor
## 0.3223dev.off()## Cairo
## 2## Compare t0, lib2 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='11' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib2t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.72, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2559 0.3369
## sample estimates:
## cor
## 0.2969
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -4.958 -1.620 -0.263 1.735 12.205
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.7276 0.1297 28.7 <2e-16 ***
## first 0.3032 0.0217 13.9 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.38
## Multiple R-squared: 0.0945, Adjusted R-squared: 0.0941
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1335 weights are ~= 1. The remaining 615 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0678 0.8520 0.9370 0.8850 0.9820 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.267 Min. : 0.696
## 1st Qu.: 3.588 1st Qu.: 3.543
## Median : 5.016 Median : 5.193
## Mean : 5.400 Mean : 5.488
## 3rd Qu.: 7.042 3rd Qu.: 7.220
## Max. :18.567 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.27
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.27
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00183004737479312 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.72, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2559 0.3369
## sample estimates:
## cor
## 0.2969dev.off()## Cairo
## 2## Compare t0, lib3 lib4
comp = exprs(expt_subset(all_norm_expt, "(batch=='12' & condition=='t0')|(batch=='34' & condition=='t0')")$expressionset)
pdf(file="figures/comparisons/interlib/lib3t0_vs_lib4t0_scatter.pdf")
sc = my_linear_scatter(comp)## [1] "Calculating correlation between the axes."
##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.31, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2475 0.3289
## sample estimates:
## cor
## 0.2888
##
## [1] "Calculating linear model between the axes"
##
## Call:
## lmrob(formula = second ~ first, data = df, method = "SMDM")
## \--> method = "SMDM"
## Residuals:
## Min 1Q Median 3Q Max
## -5.3849 -1.6005 -0.0988 1.6521 11.9325
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.8213 0.1271 30.1 <2e-16 ***
## first 0.2877 0.0212 13.5 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Robust residual standard error: 2.39
## Multiple R-squared: 0.0897, Adjusted R-squared: 0.0893
## Convergence in 6 IRWLS iterations
##
## Robustness weights:
## 1346 weights are ~= 1. The remaining 604 ones are summarized as
## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0759 0.8280 0.9340 0.8760 0.9810 0.9990
## Algorithmic parameters:
## tuning.chi1 tuning.chi2 tuning.chi3 tuning.chi4 bb tuning.psi1
## -5.0e-01 1.5e+00 NA 5.0e-01 5.0e-01 -5.0e-01
## tuning.psi2 tuning.psi3 tuning.psi4 refine.tol rel.tol solve.tol
## 1.5e+00 9.5e-01 NA 1.0e-07 1.0e-07 1.0e-07
## nResample max.it best.r.s k.fast.s k.max
## 500 50 2 1 200
## maxit.scale trace.lev mts compute.rd numpoints
## 200 0 1000 0 10
## fast.s.large.n
## 2000
## psi subsampling cov
## "lqq" "nonsingular" ".vcov.w"
## seed : int(0)
## [1] "Generating histogram of the x axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating histogram of the y axis."
## [1] "No binwidth provided, setting it to in order to have 10000 bins."
## [1] "Generating a histogram comparing the axes."
## [1] "Summarise the data."
## first second
## Min. : 0.323 Min. : 0.696
## 1st Qu.: 3.359 1st Qu.: 3.543
## Median : 5.292 Median : 5.193
## Mean : 5.408 Mean : 5.488
## 3rd Qu.: 7.054 3rd Qu.: 7.220
## Max. :18.768 Max. :17.984
## [1] "Uncorrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.33
##
## P value adjustment method: none
## [1] "Bon Ferroni corrected t test(s) between columns:"
##
## Pairwise comparisons using t tests with pooled SD
##
## data: play_all$expression and play_all$cond
##
## first
## second 0.33
##
## P value adjustment method: bonferroni
## [1] "No binwidth provided, setting it to 0.00184450670225201 in order to have 10000 bins."sc$correlation##
## Pearson's product-moment correlation
##
## data: df[, 1] and df[, 2]
## t = 13.31, df = 1948, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.2475 0.3289
## sample estimates:
## cor
## 0.2888dev.off()## Cairo
## 2