2025_exp_9_alginate_DEG

1 Overview

	batch	condition	media	treatment	induced	exp_9
SK085	1	SCFM-IPTG	SCFM	DMSO	yes	control
SK086	2	SCFM-IPTG	SCFM	DMSO	yes	control
SK087	3	SCFM-IPTG	SCFM	DMSO	yes	control
SK091	1	SCFM-Eb-IPTG	SCFM	Eb	yes	treatment
SK092	2	SCFM-Eb-IPTG	SCFM	Eb	yes	treatment
SK093	3	SCFM-Eb-IPTG	SCFM	Eb	yes	treatment

2 Project variables

##       batch    condition media treatment induced     exp_9
## SK085     1    SCFM-IPTG  SCFM      DMSO     yes   control
## SK086     2    SCFM-IPTG  SCFM      DMSO     yes   control
## SK087     3    SCFM-IPTG  SCFM      DMSO     yes   control
## SK091     1 SCFM-Eb-IPTG  SCFM        Eb     yes treatment
## SK092     2 SCFM-Eb-IPTG  SCFM        Eb     yes treatment
## SK093     3 SCFM-Eb-IPTG  SCFM        Eb     yes treatment

## [1] "The control is: SCFM-IPTG"

## [1] "The treatment is: SCFM-Eb-IPTG"

# Load raw count data
rawcounts <- read.csv("../../raw-data/htseq_SK_unfiltered_reverse_strand_count.csv", 
                      row.names = 1, header = TRUE)

# Append a prefix to the batch and run columns to turn to character class
metadata[["batch"]] <- paste0("b", metadata[["batch"]])
metadata[["run"]] <- paste0("r", metadata[["run"]])

# Subset the rawcounts dataframe to only include samples that are present in experiment
samples <- rownames(metadata)
rawcounts <- select(rawcounts, all_of(samples))

# Ensure that the order of samples in metadata matches the order of columns in rawcounts
all(rownames(metadata) == colnames(rawcounts))  # Should return TRUE if order is correct

## [1] TRUE

# Define the design formula for DESeq2, including main condition and batch effect
design_formula <- ~ exp_9 + batch

# Create DESeq2 dataset object from the filtered count matrix and metadata
dds <- DESeqDataSetFromMatrix(
  countData = rawcounts,
  colData = metadata,
  design = design_formula
)

## Warning in DESeqDataSet(se, design = design, ignoreRank): some variables in
## design formula are characters, converting to factors

3 Library size per sample

# Sum up the counts for each sample
rawcounts_sums <- colSums(rawcounts)

# Turn it into a data frame
data <- data.frame(
  Category = names(rawcounts_sums),
  Count = as.numeric(rawcounts_sums)
)

# Make sure sample order is consistent
data <- data %>%
  mutate(Category = factor(Category, levels = sort(unique(Category))))

# Plot bar chart of library sizes
p <- ggplot(data, aes(x = Category, y = Count, fill = Category)) +
  geom_bar(stat = "identity") +
  # coord_flip() # flip if too many samples to read labels easily
  labs(y = "Library size (M)", x = NULL) +  # remove x label, rename y axis
  # scale_fill_brewer(palette = "Dark2") +  # optional color palette
  scale_y_continuous(
    labels = scales::label_number(scale = 1e-6),  # show counts in millions
    limits = c(0, (max(data$Count) + 1e6))        # add a bit of space on top
  ) +
  theme(
    panel.grid = element_blank(),
    panel.background = element_blank(),
    axis.text = element_text(size = text.size, angle = 45, hjust = 1),  # slant labels
    axis.title = element_text(size = axis.text.size),
    legend.text = element_text(size = text.size),
    legend.title = element_text(size = axis.text.size),
    panel.border = element_rect(fill = NA),
    legend.position = "none",  # no legend needed since each bar is a different sample
    axis.ticks = element_line(linewidth = 0.2),
    legend.key = element_blank()
  )

# Show the plot
p

# Save the plot as a PDF with today’s date and project name
ggsave(paste0(Sys.Date(), "_library_size_", exp, "_", treatment,".vs.", control, ".pdf"), device = "pdf")

## Saving 7 x 5 in image

4 Expression distribution

# Stack rawcounts into single vector 
dat <- stack(rawcounts)

# Format expression distribution plot with the expression level and density 
expression_dist <- ggplot(dat, aes(x = log2(values+1),  fill=ind, label = ind )) + 
  geom_density(alpha = 0.5, position = "identity") + 
  ylab("Density") + 
  geom_textdensity(hjust = "ymax", vjust = 0,
           text_only = TRUE, text_smoothing = 20) + 
    theme(panel.grid = element_blank(),
        panel.background = element_blank(),
        axis.text = element_text(size = text.size),
        axis.title = element_text(size = axis.text.size),
        legend.text = element_text(size = text.size),
        legend.title = element_text(size = axis.text.size),
        panel.border = element_rect(fill = NA),
        legend.position = "none",
        axis.ticks = element_line(linewidth = 0.2),
        legend.key = element_blank()) +
  coord_cartesian(xlim = c(0, 15))

expression_dist

#save figure
cairo_pdf(paste0(Sys.Date(), "_expression_dist_", exp, "_", treatment, ".vs.", control, ".pdf"))
expression_dist
dev.off() 

## quartz_off_screen 
##                 2

5 Non-zero genes observed

# Prepare data by getting library sizes and non-zero counts
x <- colSums(rawcounts)
y <- colSums(rawcounts != 0) 
df <- data.frame(x,y)
df$ind <- rownames(df)

# Calculate axis limits
x_min <- min(df$x)
x_max <- max(df$x)
y_min <- min(df$y)
y_max <- max(df$y)

# Non-zero plot
nonzero <- ggplot(data = df, aes(x = x, y = y, color = ind, label = ind)) +
  geom_point(size = 3) +
  geom_text(hjust = -0.1, vjust = -0.1, color = "black", size = 5) +
  scale_y_continuous(limits=c(y_min, y_max)) + 
  scale_x_continuous(
    limits = c(x_min-1e3, x_max+9e5),
    labels = scales::comma_format(scale = 1e-6,  accuracy = 1)  # Format labels without decimals
  ) +
  xlab("Millions of reads") + ylab("Number of non-zero genes observed") +
  labs(color = "Sample") + 
    theme(panel.grid = element_blank(),
        panel.background = element_blank(),
        axis.text = element_text(size = text.size),
        axis.title = element_text(size = axis.text.size),
        legend.text = element_text(size = text.size),
        legend.title = element_text(size = axis.text.size),
        panel.border = element_rect(fill = NA),
        legend.position = "none",
        axis.ticks = element_line(linewidth = 0.2),
        legend.key = element_blank()) +
  coord_cartesian()

# Show nonzero plot
nonzero

# Save plot

cairo_pdf(paste0(Sys.Date(), "_non-zero_", exp, "_", treatment, ".vs.", control, ".pdf"))
nonzero
dev.off() 

## quartz_off_screen 
##                 2

6 RPKM and CPM transformations

# Load GFF annotation 
gff_annot <- load_gff_annotations("../../raw-data/paeruginosa_pa14.gff", 
                                  id_col = "gene_id", type = "gene")

## Returning a df with 16 columns and 5979 rows.

rownames(gff_annot) <- gff_annot$gene_id

head(gff_annot)

##               seqnames start  end width strand    source type score phase
## gene1650835 chromosome   483 2027  1545      + PseudoCAP gene    NA     0
## gene1650837 chromosome  2056 3159  1104      + PseudoCAP gene    NA     0
## gene1650839 chromosome  3169 4278  1110      + PseudoCAP gene    NA     0
## gene1650841 chromosome  4275 6695  2421      + PseudoCAP gene    NA     0
## gene1650843 chromosome  7018 7791   774      - PseudoCAP gene    NA     0
## gene1650845 chromosome  7803 8339   537      - PseudoCAP gene    NA     0
##                 gene_id Name         Dbxref      Alias       name Parent locus
## gene1650835 gene1650835      GeneID:4384099 PA14_00010       dnaA         <NA>
## gene1650837 gene1650837      GeneID:4384100 PA14_00020       dnaN         <NA>
## gene1650839 gene1650839      GeneID:4384101 PA14_00030       recF         <NA>
## gene1650841 gene1650841      GeneID:4384102 PA14_00050       gyrB         <NA>
## gene1650843 gene1650843      GeneID:4385186 PA14_00060 PA14_00060         <NA>
## gene1650845 gene1650845      GeneID:4385187 PA14_00070 PA14_00070         <NA>

head(rawcounts)

##             SK085 SK086 SK087 SK091 SK092 SK093
## gene1650835  1579  1694  1158  1415  1704  1788
## gene1650837  1732  2280  1356  2089  2657  2276
## gene1650839   377   454   293   413   373   458
## gene1650841  1914  2500  1745  2529  2710  3079
## gene1650843   292   389   236   370   297   365
## gene1650845   177   180   124   175   156   151

dim(gff_annot)

## [1] 5979   16

dim(rawcounts)

## [1] 5979    6

summary(rownames(gff_annot) == rownames(rawcounts))

##    Mode   FALSE    TRUE 
## logical     932    5047

gff_annot <- gff_annot[rownames(rawcounts),]
summary(rownames(gff_annot) == rownames(rawcounts))

##    Mode    TRUE 
## logical    5979

# Create an expression object with counts, metadata, and gene info
expt <- create_expt(count_dataframe = rawcounts,
                    metadata = metadata, 
                    gene_info = gff_annot)

## Reading the sample metadata.

## Did not find the column: sampleid.

## The rownames do not appear numeric, using them.

## The sample definitions comprises: 6 rows(samples) and 8 columns(metadata fields).

## Matched 5979 annotations and counts.

## Bringing together the count matrix and gene information.

## Some annotations were lost in merging, setting them to 'undefined'.

## Saving the expressionset to 'expt.rda'.

## The final expressionset has 5979 features and 6 samples.

# Normalize counts log2 CPM
cpm_normalized_exp <- normalize_expt(expt, 
                                     transform = "log2", 
                                     norm = "raw", 
                                     convert = "cpm", 
                                     row_min = "10")

## transform_counts: Found 259 values equal to 0, adding 1 to the matrix.

####### Calculate RPKM ########

# Grab gene lengths from GFF (should be in base pairs)
gene_lengths <- gff_annot$width  
names(gene_lengths) <- rownames(gff_annot)  # set gene names

# Make sure gene_lengths are in the same order as rawcounts rows
gene_lengths <- gene_lengths[rownames(rawcounts)]

# Make DGEList object for edgeR
dge <- DGEList(counts = rawcounts)

# Calculate RPKM (reads per kilobase per million)
rpkm <- rpkm(dge, gene.length = gene_lengths)

# Wrap RPKM in an experiment object too
expt_rpkm <- create_expt(count_dataframe = rpkm,
                         metadata = metadata, 
                         gene_info = gff_annot)

## Reading the sample metadata.

## Did not find the column: sampleid.

## The rownames do not appear numeric, using them.

## The sample definitions comprises: 6 rows(samples) and 8 columns(metadata fields).

## Matched 5979 annotations and counts.

## Bringing together the count matrix and gene information.

## Some annotations were lost in merging, setting them to 'undefined'.

## Saving the expressionset to 'expt.rda'.

## The final expressionset has 5979 features and 6 samples.

# log2 transform the RPKM (again, no normalization beyond log)
rpkm_normalized_exp <- normalize_expt(expt_rpkm, 
                                      transform = "log2", 
                                      norm = "raw")

## transform_counts: Found 259 values equal to 0, adding 1 to the matrix.

# Get the actual expression matrices from the experiment objects
rpkm <- exprs(rpkm_normalized_exp)
cpm <- exprs(cpm_normalized_exp)

# Merge CPM and RPKM matrices with gene annotation info
cpm_annotated <- merge(gff_annot, cpm, by = 0)
rpkm_annotated <- merge(gff_annot, rpkm, by = 0)

6.1 Heatmap of alginate genes

This generates a small heatmap of the alginate genes across samples to visualize expression.

# Loop through CPM and RPKM to plot both heatmaps
for (type in c("cpm", "rpkm")) {
  
  # Define gene list
  alg_operon_genes <- c(
    "algA", "algF", "algJ", "algI", "algL", "algX", "algG", "algE",
    "algK", "alg44", "alg8", "algD", "mucC", "mucB", "algU", "algW",
    "algP", "algQ", "algR", "algZ", "algC", "algB"
  )

  # Get the right data per type of transformation
  data_annot <- get(paste0(type, "_annotated"))
  
  # Fix rownames using gene name (make unique)
  rownames(data_annot) <- make.names(data_annot$name, unique = TRUE)
  
  # Drop undesired metadata columns
  cols_to_exclude <- c("Row.names","seqnames","start","end","width",
                       "strand","source","type","score","phase",
                       "gene_id","Name","Dbxref","Alias","name",
                       "Parent","locus")
  keep <- !names(data_annot) %in% cols_to_exclude
  expr_data <- data_annot[, keep]
  
  # Extract matrix for alg genes
  alg_genes <- expr_data[alg_operon_genes, , drop = FALSE]
  mat <- alg_genes - rowMeans(alg_genes)
  
  # Pull sample annotations
  anno <- as.data.frame(metadata[, c("condition", "induced")])
  colnames(anno) <- c("Condition", "Induced")
  
  # Color settings for annotation
 # ann_colors <- list(
 #  Condition = c("LB-DMSO" = "#D95F02", "LB-IPTG" = "#66A61E"),
 #  Induced = c("no" = "lightgrey", "yes" = "gray27")
 # )
  
  # Scale data by row (Z-score per gene)
  scaled_mat <- t(scale(t(mat)))

  # Build heatmap
  ht <- Heatmap(
    scaled_mat,
    name = "Z-score",
    top_annotation = HeatmapAnnotation(df = anno#, 
                                       #col = ann_colors
                                       ),
    cluster_rows = TRUE,
    cluster_columns = TRUE,
    row_names_side = "left",
    column_names_side = "bottom",
    column_names_rot = 45,
    row_names_gp = gpar(fontsize = 10),
    column_names_gp = gpar(fontsize = 10),
    show_row_dend = TRUE,
    show_column_dend = TRUE
  )
  
  # Draw and save
  draw(ht)
  
  cairo_pdf(paste0(Sys.Date(), "_alg_heatmap_", type, "_", exp, "_", treatment, ".vs.", control, ".pdf"))
  draw(ht)
  dev.off()
}

7 DESeq

# Create DESeqDataSet
dds <- DESeqDataSetFromMatrix(
  countData = rawcounts,
  colData = metadata,
  design = design_formula
)

## Warning in DESeqDataSet(se, design = design, ignoreRank): some variables in
## design formula are characters, converting to factors

# Check number of genes before filtering
print("Number of genes before filtering:")

## [1] "Number of genes before filtering:"

nrow(dds)

## [1] 5979

# Filter: keep genes with counts >=10 in at least 2 samples
print("Filtering genes with counts >=10 in at least 2 samples")

## [1] "Filtering genes with counts >=10 in at least 2 samples"

keep <- rowSums(counts(dds) >= 10) >= 2 
dds_filtered <- dds[keep, ]

# Check number of genes after filtering
print("Number of genes after filtering:")

## [1] "Number of genes after filtering:"

nrow(dds_filtered)

## [1] 5740

# Run DESeq on the filtered dataset
dds_final <- DESeq(dds_filtered)

## estimating size factors

## estimating dispersions

## gene-wise dispersion estimates

## mean-dispersion relationship

## final dispersion estimates

## fitting model and testing

# Show available results
resultsNames(dds_final)

## [1] "Intercept"                  "exp_9_treatment_vs_control"
## [3] "batch_b2_vs_b1"             "batch_b3_vs_b1"

# Plot dispersion estimates
plotDispEsts(dds_final)

8 Save DE tables

res <- results(dds_final, contrast = c("exp_9", "treatment", "control"))
resdf <- as.data.frame(res)
resdf$gene_id <- rownames(resdf)

cpm <- as.data.frame(cpm)
rpkm <- as.data.frame(rpkm)

# append type of value to the dataframe   
colnames(rpkm) <- paste0(colnames(rpkm), "_rpkm")
colnames(cpm) <- paste0(colnames(cpm), "_cpm")
rawcounts2 <- rawcounts
colnames(rawcounts2) <- paste0(colnames(rawcounts2), "_rawcounts")

# ensure column gene_id is there
rpkm$gene_id <- rownames(rpkm)
cpm$gene_id <- rownames(cpm)
rawcounts2$gene_id <- rownames(rawcounts2)

gff_annot <- gff_annot[rownames(resdf),]

dim(resdf)

## [1] 5740    7

# merge with siggenes
resdf <- full_join(resdf, gff_annot, by = "gene_id", copy = FALSE)
head(resdf)

##    baseMean log2FoldChange     lfcSE       stat     pvalue      padj
## 1 1534.7307    -0.06737316 0.1174606 -0.5735810 0.56625137 0.8432258
## 2 2017.6432     0.18654789 0.1044253  1.7864237 0.07403068 0.2912723
## 3  389.1677    -0.06325250 0.1600921 -0.3951007 0.69276858 0.9068755
## 4 2364.4787     0.21904612 0.1033496  2.1194668 0.03405103 0.1708277
## 5  319.9061    -0.03352170 0.1804008 -0.1858180 0.85258751 0.9593294
## 6  159.5413    -0.20465001 0.2182824 -0.9375471 0.34847723 0.6855253
##       gene_id   seqnames start  end width strand    source type score phase
## 1 gene1650835 chromosome   483 2027  1545      + PseudoCAP gene    NA     0
## 2 gene1650837 chromosome  2056 3159  1104      + PseudoCAP gene    NA     0
## 3 gene1650839 chromosome  3169 4278  1110      + PseudoCAP gene    NA     0
## 4 gene1650841 chromosome  4275 6695  2421      + PseudoCAP gene    NA     0
## 5 gene1650843 chromosome  7018 7791   774      - PseudoCAP gene    NA     0
## 6 gene1650845 chromosome  7803 8339   537      - PseudoCAP gene    NA     0
##   Name         Dbxref      Alias       name Parent locus
## 1      GeneID:4384099 PA14_00010       dnaA         <NA>
## 2      GeneID:4384100 PA14_00020       dnaN         <NA>
## 3      GeneID:4384101 PA14_00030       recF         <NA>
## 4      GeneID:4384102 PA14_00050       gyrB         <NA>
## 5      GeneID:4385186 PA14_00060 PA14_00060         <NA>
## 6      GeneID:4385187 PA14_00070 PA14_00070         <NA>

pa_go <- read.csv("../../raw-data/208963_microbes.online.csv", header = TRUE)
head(pa_go)

##   locusId   accession        GI scaffoldId start stop strand      Alias
## 1 2194572 YP_788156.1 116053721       4582   483 2027      + PA14_00010
## 2 2194573 YP_788157.1 116053722       4582  2056 3159      + PA14_00020
## 3 2194574 YP_788158.1 116053723       4582  3169 4278      + PA14_00030
## 4 2194575 YP_788159.1 116053724       4582  4275 6695      + PA14_00050
## 5 2194576 YP_788160.1 116053725       4582  7791 7018      - PA14_00060
## 6 2194577 YP_788161.1 116053726       4582  8339 7803      - PA14_00070
##         name                                                  desc     COG
## 1       dnaA chromosomal replication initiator protein DnaA (NCBI)  COG593
## 2       dnaN                 DNA polymerase III, beta chain (NCBI)  COG592
## 3       recF        DNA replication and repair protein RecF (NCBI) COG1195
## 4       gyrB                           DNA gyrase subunit B (NCBI)  COG187
## 5 PA14_00060                       putative acyltransferase (NCBI)  COG204
## 6 PA14_00070                putative histidinol-phosphatase (NCBI)  COG241
##   COGFun
## 1      L
## 2      L
## 3      L
## 4      L
## 5      I
## 6      E
##                                                                    COGDesc
## 1                            ATPase involved in DNA replication initiation
## 2                      DNA polymerase sliding clamp subunit (PCNA homolog)
## 3                         Recombinational DNA repair ATPase (RecF pathway)
## 4 Type IIA topoisomerase (DNA gyrase/topo II, topoisomerase IV), B subunit
## 5                           1-acyl-sn-glycerol-3-phosphate acyltransferase
## 6                          Histidinol phosphatase and related phosphatases
##                                                                                  TIGRFam
## 1                        TIGR00362 chromosomal replication initiator protein DnaA [dnaA]
## 2                                      TIGR00663 DNA polymerase III, beta subunit [dnaN]
## 3                               TIGR00611 DNA replication and repair protein RecF [recF]
## 4                                                 TIGR01059 DNA gyrase, B subunit [gyrB]
## 5                                                                                       
## 6 TIGR01656 histidinol-phosphate phosphatase domain,TIGR01662 HAD hydrolase, family IIIA
##                                                   TIGRRoles
## 1 DNA metabolism:DNA replication, recombination, and repair
## 2 DNA metabolism:DNA replication, recombination, and repair
## 3 DNA metabolism:DNA replication, recombination, and repair
## 4 DNA metabolism:DNA replication, recombination, and repair
## 5                                                          
## 6           Unknown function:Enzymes of unknown specificity
##                                                                                                                                                                                           GO
## 1                                                                                                                                     GO:0006270,GO:0006275,GO:0003688,GO:0017111,GO:0005524
## 2 GO:0006260,GO:0003677,GO:0003893,GO:0008408,GO:0016449,GO:0019984,GO:0003889,GO:0003894,GO:0015999,GO:0016450,GO:0003890,GO:0003895,GO:0016000,GO:0016451,GO:0003891,GO:0016448,GO:0016452
## 3                                                                                                                                     GO:0006281,GO:0005694,GO:0005524,GO:0017111,GO:0003697
## 4                                                                                                                                     GO:0006304,GO:0006265,GO:0005694,GO:0003918,GO:0005524
## 5                                                                                                                                                                      GO:0008152,GO:0003841
## 6                                                                                                                                                                      GO:0000105,GO:0004401
##         EC                                        ECDesc
## 1                                                       
## 2  2.7.7.7                  DNA-directed DNA polymerase.
## 3                                                       
## 4 5.99.1.3          DNA topoisomerase (ATP-hydrolyzing).
## 5 2.3.1.51 1-acylglycerol-3-phosphate O-acyltransferase.
## 6  3.1.3.-

col_to_remove <- intersect(colnames(pa_go),colnames(resdf))
col_to_remove <- setdiff(col_to_remove, "Alias") # returns all except the second vector

pa_go <- pa_go %>% select(-all_of(col_to_remove))

resdf <- full_join(resdf, rpkm, by = "gene_id", copy = FALSE)
resdf <- full_join(resdf, cpm, by = "gene_id", copy = FALSE)
resdf <- full_join(resdf, rawcounts2, by = "gene_id", copy = FALSE)
#resdf <- full_join(resdf, pa_go, by = "Alias", copy = FALSE)
resdf <- merge(resdf, pa_go, by = "Alias", copy = FALSE)

head(resdf)

##        Alias  baseMean log2FoldChange     lfcSE       stat     pvalue      padj
## 1 PA14_00010 1534.7307    -0.06737316 0.1174606 -0.5735810 0.56625137 0.8432258
## 2 PA14_00020 2017.6432     0.18654789 0.1044253  1.7864237 0.07403068 0.2912723
## 3 PA14_00030  389.1677    -0.06325250 0.1600921 -0.3951007 0.69276858 0.9068755
## 4 PA14_00050 2364.4787     0.21904612 0.1033496  2.1194668 0.03405103 0.1708277
## 5 PA14_00060  319.9061    -0.03352170 0.1804008 -0.1858180 0.85258751 0.9593294
## 6 PA14_00070  159.5413    -0.20465001 0.2182824 -0.9375471 0.34847723 0.6855253
##       gene_id   seqnames start  end width strand    source type score phase
## 1 gene1650835 chromosome   483 2027  1545      + PseudoCAP gene    NA     0
## 2 gene1650837 chromosome  2056 3159  1104      + PseudoCAP gene    NA     0
## 3 gene1650839 chromosome  3169 4278  1110      + PseudoCAP gene    NA     0
## 4 gene1650841 chromosome  4275 6695  2421      + PseudoCAP gene    NA     0
## 5 gene1650843 chromosome  7018 7791   774      - PseudoCAP gene    NA     0
## 6 gene1650845 chromosome  7803 8339   537      - PseudoCAP gene    NA     0
##   Name         Dbxref       name Parent locus SK085_rpkm SK086_rpkm SK087_rpkm
## 1      GeneID:4384099       dnaA         <NA>   7.989449   7.833958   7.830711
## 2      GeneID:4384100       dnaN         <NA>   8.605764   8.744455   8.540827
## 3      GeneID:4384101       recF         <NA>   6.411482   6.421924   6.336705
## 4      GeneID:4384102       gyrB         <NA>   7.620694   7.747847   7.774559
## 5      GeneID:4385186 PA14_00060         <NA>   6.561349   6.716019   6.542347
## 6      GeneID:4385187 PA14_00070         <NA>   6.368749   6.138494   6.146263
##   SK091_rpkm SK092_rpkm SK093_rpkm SK085_cpm SK086_cpm SK087_cpm SK091_cpm
## 1   7.699884   7.849157   7.827288  8.615052  8.459333  8.456080  8.325041
## 2   8.743177   8.971503   8.657519  8.748155  8.886879  8.683202  8.885600
## 3   6.410449   6.148735   6.350827  6.560362  6.570815  6.485495  6.559326
## 4   7.888797   7.870437   7.962835  8.891988  9.019505  9.046289  9.160823
## 5   6.768222   6.337647   6.541318  6.196208  6.350426  6.177265  6.402487
## 6   6.221629   5.941855   5.805831  5.486717  5.259038  5.266713  5.341196
##   SK092_cpm SK093_cpm SK085_rawcounts SK086_rawcounts SK087_rawcounts
## 1  8.474555  8.452652            1579            1694            1158
## 2  9.113973  8.799923            1732            2280            1356
## 3  6.297278  6.499634             377             454             293
## 4  9.142417  9.235040            1914            2500            1745
## 5  5.973252  6.176239             292             389             236
## 6  5.064943  4.930891             177             180             124
##   SK091_rawcounts SK092_rawcounts SK093_rawcounts locusId   accession        GI
## 1            1415            1704            1788 2194572 YP_788156.1 116053721
## 2            2089            2657            2276 2194573 YP_788157.1 116053722
## 3             413             373             458 2194574 YP_788158.1 116053723
## 4            2529            2710            3079 2194575 YP_788159.1 116053724
## 5             370             297             365 2194576 YP_788160.1 116053725
## 6             175             156             151 2194577 YP_788161.1 116053726
##   scaffoldId stop                                                  desc     COG
## 1       4582 2027 chromosomal replication initiator protein DnaA (NCBI)  COG593
## 2       4582 3159                 DNA polymerase III, beta chain (NCBI)  COG592
## 3       4582 4278        DNA replication and repair protein RecF (NCBI) COG1195
## 4       4582 6695                           DNA gyrase subunit B (NCBI)  COG187
## 5       4582 7018                       putative acyltransferase (NCBI)  COG204
## 6       4582 7803                putative histidinol-phosphatase (NCBI)  COG241
##   COGFun
## 1      L
## 2      L
## 3      L
## 4      L
## 5      I
## 6      E
##                                                                    COGDesc
## 1                            ATPase involved in DNA replication initiation
## 2                      DNA polymerase sliding clamp subunit (PCNA homolog)
## 3                         Recombinational DNA repair ATPase (RecF pathway)
## 4 Type IIA topoisomerase (DNA gyrase/topo II, topoisomerase IV), B subunit
## 5                           1-acyl-sn-glycerol-3-phosphate acyltransferase
## 6                          Histidinol phosphatase and related phosphatases
##                                                                                  TIGRFam
## 1                        TIGR00362 chromosomal replication initiator protein DnaA [dnaA]
## 2                                      TIGR00663 DNA polymerase III, beta subunit [dnaN]
## 3                               TIGR00611 DNA replication and repair protein RecF [recF]
## 4                                                 TIGR01059 DNA gyrase, B subunit [gyrB]
## 5                                                                                       
## 6 TIGR01656 histidinol-phosphate phosphatase domain,TIGR01662 HAD hydrolase, family IIIA
##                                                   TIGRRoles
## 1 DNA metabolism:DNA replication, recombination, and repair
## 2 DNA metabolism:DNA replication, recombination, and repair
## 3 DNA metabolism:DNA replication, recombination, and repair
## 4 DNA metabolism:DNA replication, recombination, and repair
## 5                                                          
## 6           Unknown function:Enzymes of unknown specificity
##                                                                                                                                                                                           GO
## 1                                                                                                                                     GO:0006270,GO:0006275,GO:0003688,GO:0017111,GO:0005524
## 2 GO:0006260,GO:0003677,GO:0003893,GO:0008408,GO:0016449,GO:0019984,GO:0003889,GO:0003894,GO:0015999,GO:0016450,GO:0003890,GO:0003895,GO:0016000,GO:0016451,GO:0003891,GO:0016448,GO:0016452
## 3                                                                                                                                     GO:0006281,GO:0005694,GO:0005524,GO:0017111,GO:0003697
## 4                                                                                                                                     GO:0006304,GO:0006265,GO:0005694,GO:0003918,GO:0005524
## 5                                                                                                                                                                      GO:0008152,GO:0003841
## 6                                                                                                                                                                      GO:0000105,GO:0004401
##         EC                                        ECDesc
## 1                                                       
## 2  2.7.7.7                  DNA-directed DNA polymerase.
## 3                                                       
## 4 5.99.1.3          DNA topoisomerase (ATP-hydrolyzing).
## 5 2.3.1.51 1-acylglycerol-3-phosphate O-acyltransferase.
## 6  3.1.3.-

write.csv(resdf, paste0(Sys.Date() , "_results_",exp, "_", treatment,".vs.", control,".csv"), row.names = FALSE)

9 Save DEG list

# filter for significantly differentially expressed genes (LFC >= 3, padj < pval_cutoff)
dim(resdf)

## [1] 5733   54

head(resdf)

##        Alias  baseMean log2FoldChange     lfcSE       stat     pvalue      padj
## 1 PA14_00010 1534.7307    -0.06737316 0.1174606 -0.5735810 0.56625137 0.8432258
## 2 PA14_00020 2017.6432     0.18654789 0.1044253  1.7864237 0.07403068 0.2912723
## 3 PA14_00030  389.1677    -0.06325250 0.1600921 -0.3951007 0.69276858 0.9068755
## 4 PA14_00050 2364.4787     0.21904612 0.1033496  2.1194668 0.03405103 0.1708277
## 5 PA14_00060  319.9061    -0.03352170 0.1804008 -0.1858180 0.85258751 0.9593294
## 6 PA14_00070  159.5413    -0.20465001 0.2182824 -0.9375471 0.34847723 0.6855253
##       gene_id   seqnames start  end width strand    source type score phase
## 1 gene1650835 chromosome   483 2027  1545      + PseudoCAP gene    NA     0
## 2 gene1650837 chromosome  2056 3159  1104      + PseudoCAP gene    NA     0
## 3 gene1650839 chromosome  3169 4278  1110      + PseudoCAP gene    NA     0
## 4 gene1650841 chromosome  4275 6695  2421      + PseudoCAP gene    NA     0
## 5 gene1650843 chromosome  7018 7791   774      - PseudoCAP gene    NA     0
## 6 gene1650845 chromosome  7803 8339   537      - PseudoCAP gene    NA     0
##   Name         Dbxref       name Parent locus SK085_rpkm SK086_rpkm SK087_rpkm
## 1      GeneID:4384099       dnaA         <NA>   7.989449   7.833958   7.830711
## 2      GeneID:4384100       dnaN         <NA>   8.605764   8.744455   8.540827
## 3      GeneID:4384101       recF         <NA>   6.411482   6.421924   6.336705
## 4      GeneID:4384102       gyrB         <NA>   7.620694   7.747847   7.774559
## 5      GeneID:4385186 PA14_00060         <NA>   6.561349   6.716019   6.542347
## 6      GeneID:4385187 PA14_00070         <NA>   6.368749   6.138494   6.146263
##   SK091_rpkm SK092_rpkm SK093_rpkm SK085_cpm SK086_cpm SK087_cpm SK091_cpm
## 1   7.699884   7.849157   7.827288  8.615052  8.459333  8.456080  8.325041
## 2   8.743177   8.971503   8.657519  8.748155  8.886879  8.683202  8.885600
## 3   6.410449   6.148735   6.350827  6.560362  6.570815  6.485495  6.559326
## 4   7.888797   7.870437   7.962835  8.891988  9.019505  9.046289  9.160823
## 5   6.768222   6.337647   6.541318  6.196208  6.350426  6.177265  6.402487
## 6   6.221629   5.941855   5.805831  5.486717  5.259038  5.266713  5.341196
##   SK092_cpm SK093_cpm SK085_rawcounts SK086_rawcounts SK087_rawcounts
## 1  8.474555  8.452652            1579            1694            1158
## 2  9.113973  8.799923            1732            2280            1356
## 3  6.297278  6.499634             377             454             293
## 4  9.142417  9.235040            1914            2500            1745
## 5  5.973252  6.176239             292             389             236
## 6  5.064943  4.930891             177             180             124
##   SK091_rawcounts SK092_rawcounts SK093_rawcounts locusId   accession        GI
## 1            1415            1704            1788 2194572 YP_788156.1 116053721
## 2            2089            2657            2276 2194573 YP_788157.1 116053722
## 3             413             373             458 2194574 YP_788158.1 116053723
## 4            2529            2710            3079 2194575 YP_788159.1 116053724
## 5             370             297             365 2194576 YP_788160.1 116053725
## 6             175             156             151 2194577 YP_788161.1 116053726
##   scaffoldId stop                                                  desc     COG
## 1       4582 2027 chromosomal replication initiator protein DnaA (NCBI)  COG593
## 2       4582 3159                 DNA polymerase III, beta chain (NCBI)  COG592
## 3       4582 4278        DNA replication and repair protein RecF (NCBI) COG1195
## 4       4582 6695                           DNA gyrase subunit B (NCBI)  COG187
## 5       4582 7018                       putative acyltransferase (NCBI)  COG204
## 6       4582 7803                putative histidinol-phosphatase (NCBI)  COG241
##   COGFun
## 1      L
## 2      L
## 3      L
## 4      L
## 5      I
## 6      E
##                                                                    COGDesc
## 1                            ATPase involved in DNA replication initiation
## 2                      DNA polymerase sliding clamp subunit (PCNA homolog)
## 3                         Recombinational DNA repair ATPase (RecF pathway)
## 4 Type IIA topoisomerase (DNA gyrase/topo II, topoisomerase IV), B subunit
## 5                           1-acyl-sn-glycerol-3-phosphate acyltransferase
## 6                          Histidinol phosphatase and related phosphatases
##                                                                                  TIGRFam
## 1                        TIGR00362 chromosomal replication initiator protein DnaA [dnaA]
## 2                                      TIGR00663 DNA polymerase III, beta subunit [dnaN]
## 3                               TIGR00611 DNA replication and repair protein RecF [recF]
## 4                                                 TIGR01059 DNA gyrase, B subunit [gyrB]
## 5                                                                                       
## 6 TIGR01656 histidinol-phosphate phosphatase domain,TIGR01662 HAD hydrolase, family IIIA
##                                                   TIGRRoles
## 1 DNA metabolism:DNA replication, recombination, and repair
## 2 DNA metabolism:DNA replication, recombination, and repair
## 3 DNA metabolism:DNA replication, recombination, and repair
## 4 DNA metabolism:DNA replication, recombination, and repair
## 5                                                          
## 6           Unknown function:Enzymes of unknown specificity
##                                                                                                                                                                                           GO
## 1                                                                                                                                     GO:0006270,GO:0006275,GO:0003688,GO:0017111,GO:0005524
## 2 GO:0006260,GO:0003677,GO:0003893,GO:0008408,GO:0016449,GO:0019984,GO:0003889,GO:0003894,GO:0015999,GO:0016450,GO:0003890,GO:0003895,GO:0016000,GO:0016451,GO:0003891,GO:0016448,GO:0016452
## 3                                                                                                                                     GO:0006281,GO:0005694,GO:0005524,GO:0017111,GO:0003697
## 4                                                                                                                                     GO:0006304,GO:0006265,GO:0005694,GO:0003918,GO:0005524
## 5                                                                                                                                                                      GO:0008152,GO:0003841
## 6                                                                                                                                                                      GO:0000105,GO:0004401
##         EC                                        ECDesc
## 1                                                       
## 2  2.7.7.7                  DNA-directed DNA polymerase.
## 3                                                       
## 4 5.99.1.3          DNA topoisomerase (ATP-hydrolyzing).
## 5 2.3.1.51 1-acylglycerol-3-phosphate O-acyltransferase.
## 6  3.1.3.-

summary(is.na(resdf$log2FoldChange))

##    Mode   FALSE 
## logical    5733

summary(is.na(resdf$padj))

##    Mode   FALSE    TRUE 
## logical    5067     666

sig_genes <- resdf[abs(resdf$log2FoldChange) >= logfc_cutoff & resdf$padj < pval_cutoff, ]  %>% drop_na(log2FoldChange)
sig_genes$direction <- ifelse(sig_genes$log2FoldChange > 0, "up", "down")

summary(is.na(sig_genes$log2FoldChange))

##    Mode   FALSE 
## logical      18

summary(is.na(sig_genes$padj))

##    Mode   FALSE 
## logical      18

dim(sig_genes)

## [1] 18 55

head(sig_genes)

##           Alias   baseMean log2FoldChange     lfcSE      stat        pvalue
## 289  PA14_03670   65.70098      -2.037475 0.3462280 -5.884778  3.985890e-09
## 291  PA14_03700  194.97456      -2.141479 0.2330573 -9.188637  3.978521e-20
## 292  PA14_03710  151.29792      -2.852538 0.4960608 -5.750380  8.904283e-09
## 1076 PA14_13810   32.02238       2.049227 0.5190659  3.947914  7.883506e-05
## 1434 PA14_18380 2479.32913       2.981049 0.1365352 21.833550 1.114212e-105
## 1435 PA14_18410  470.42766       2.537818 0.1600833 15.853110  1.337914e-56
##               padj     gene_id   seqnames   start     end width strand
## 289   1.630358e-07 gene1651425 chromosome  329086  329955   870      -
## 291   7.473725e-18 gene1651429 chromosome  330946  331944   999      -
## 292   3.474040e-07 gene1651431 chromosome  332121  332303   183      -
## 1076  1.249536e-03 gene1653045 chromosome 1188951 1189691   741      +
## 1434 5.651285e-102 gene1653783 chromosome 1578657 1580102  1446      -
## 1435  1.696476e-53 gene1653785 chromosome 1580299 1580949   651      -
##         source type score phase Name         Dbxref       name Parent locus
## 289  PseudoCAP gene    NA     0      GeneID:4383686       cysW         <NA>
## 291  PseudoCAP gene    NA     0      GeneID:4383688        sbp         <NA>
## 292  PseudoCAP gene    NA     0      GeneID:4383689 PA14_03710         <NA>
## 1076 PseudoCAP gene    NA     0      GeneID:4381917       narJ         <NA>
## 1434 PseudoCAP gene    NA     0      GeneID:4381515       algA         <NA>
## 1435 PseudoCAP gene    NA     0      GeneID:4381514       algF         <NA>
##      SK085_rpkm SK086_rpkm SK087_rpkm SK091_rpkm SK092_rpkm SK093_rpkm
## 289    4.622323   5.008864   4.932872   2.801752   2.803576   3.218705
## 291    6.259635   5.919397   6.432344   3.647772   4.055529   4.566050
## 292    8.533314   8.004901   8.509954   5.544729   4.346073   6.536374
## 1076   1.877372   2.027034   2.832187   4.481747   3.709249   3.904480
## 1434   6.336653   6.148166   6.904321   9.704073   8.980128   9.532922
## 1435   5.850239   5.304129   5.703139   8.450330   7.675014   8.188973
##      SK085_cpm SK086_cpm SK087_cpm SK091_cpm SK092_cpm SK093_cpm
## 289   4.430136  4.814631  4.739000  2.631429  2.633215  3.040765
## 291   6.258210  5.917977  6.430917  3.646444  4.054172  4.564668
## 292   6.100510  5.579675  6.077430  3.226419  2.182301  4.154063
## 1076  1.576025  1.713284  2.468858  4.071687  3.314842  3.505312
## 1434  6.863205  6.673946  7.432669 10.235607  9.511314 10.064389
## 1435  5.244313  4.704302  5.098638  7.833269  7.059523  7.572350
##      SK085_rawcounts SK086_rawcounts SK087_rawcounts SK091_rawcounts
## 289               83             131              85              23
## 291              305             287             282              51
## 292              273             226             220              37
## 1076               8              11              15              70
## 1434             466             488             568            5332
## 1435             149             121             110            1005
##      SK092_rawcounts SK093_rawcounts locusId   accession        GI scaffoldId
## 289               25              37 2194866 YP_788450.1 116054008       4582
## 291               75             116 2194868 YP_788452.1 116054010       4582
## 292               17              86 2194869 YP_788453.1 116054011       4582
## 1076              43              53 2195664 YP_789248.1 116051909       4582
## 1434            3501            5475 2196030 YP_789614.1 116051549       4582
## 1435             636             969 2196031 YP_789615.1 116051548       4582
##         stop
## 289   329086
## 291   330946
## 292   332121
## 1076 1189691
## 1434 1578657
## 1435 1580299
##                                                                              desc
## 289                                         sulfate transport protein CysW (NCBI)
## 291                                      sulfate-binding protein precursor (NCBI)
## 292                                                   hypothetical protein (NCBI)
## 1076                             respiratory nitrate reductase delta chain (NCBI)
## 1434 mannose-1-phosphate guanylyltransferase/mannose-6-phosphate isomerase (NCBI)
## 1435                                          alginate O-acetyltransferase (NCBI)
##          COG COGFun                                                  COGDesc
## 289  COG4208      P    ABC-type sulfate transport system, permease component
## 291  COG1613      P ABC-type sulfate transport system, periplasmic component
## 292  COG5583      S                            Uncharacterized small protein
## 1076 COG2180      C                          Nitrate reductase delta subunit
## 1434  COG836      M                  Mannose-1-phosphate guanylyltransferase
## 1435                                                                        
##                                                                                                                  TIGRFam
## 289  TIGR00969 sulfate ABC transporter, permease protein,TIGR02140 sulfate ABC transporter, permease protein CysW [cysW]
## 291                                                           TIGR00971 sulfate ABC transporter, sulfate-binding protein
## 292                                                                                                                     
## 1076                                           TIGR00684 nitrate reductase molybdenum cofactor assembly chaperone [narJ]
## 1434                                     TIGR01479 mannose-1-phosphate guanylyltransferase/mannose-6-phosphate isomerase
## 1435                                                                                                                    
##                                                                                          TIGRRoles
## 289                                                          Transport and binding proteins:Anions
## 291                                                          Transport and binding proteins:Anions
## 292                                                                                               
## 1076                                                Protein fate:Protein folding and stabilization
## 1434 Cell envelope:Biosynthesis and degradation of surface polysaccharides and lipopolysaccharides
## 1435                                                                                              
##                                                          GO       EC
## 289  GO:0008272,GO:0016021,GO:0009276,GO:0015116,GO:0015563         
## 291             GO:0008272,GO:0030288,GO:0008271,GO:0015419         
## 292                                                                 
## 1076                       GO:0006118,GO:0009325,GO:0008940 1.7.99.4
## 1434                                  GO:0009103,GO:0008928 2.7.7.22
## 1435                                                                
##                                              ECDesc direction
## 289                                                      down
## 291                                                      down
## 292                                                      down
## 1076                             Nitrate reductase.        up
## 1434 Mannose-1-phosphate guanylyltransferase (GDP).        up
## 1435                                                       up

write.csv(sig_genes, paste0(Sys.Date(),"_significantgenes_", exp, "_", treatment,".vs.", control,".csv"), row.names = FALSE)

10 PCA plot

vst <- DESeqDataSetFromMatrix(countData = rawcounts, colData = metadata, design = ~ exp_9 + batch)

## Warning in DESeqDataSet(se, design = design, ignoreRank): some variables in
## design formula are characters, converting to factors

# Variance stabilizing transformations
vsd <- vst(vst, blind = TRUE)

# PCA plot
pcaData <- plotPCA(vsd, intgroup = c("exp_9", "batch", "condition"), returnData = TRUE)

## using ntop=500 top features by variance

percentVar <- round(100 * attr(pcaData, "percentVar"))

# Calculate the maximum absolute value among PC1 and PC2
max_abs_val <- max(c(abs(min(pcaData$PC1)), abs(min(pcaData$PC2)), abs(max(pcaData$PC1)), abs(max(pcaData$PC2))))

axis_size <- 10
theme_set(theme_minimal(base_family = "Arial"))

pca <- ggplot(pcaData, aes(PC1, PC2, color = condition, shape = condition)) +
  geom_point(size = 3) +
  xlab(paste0("PC1: ", percentVar[1], "% variance")) +
  ylab(paste0("PC2: ", percentVar[2], "% variance")) +
  scale_colour_brewer(palette = "Dark2") +
  scale_shape_manual(values = c(1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)) +
  coord_fixed(ratio = 1, xlim = c(-max_abs_val, max_abs_val), ylim = c(-max_abs_val, max_abs_val)) +
  labs(color = "condition", shape = "condition") +
  guides(shape = guide_legend(title = "Condition"),  # Adjust size of shape legend
    color = guide_legend(title = "Condition")) +
  theme(
    panel.grid = element_blank(),
    panel.background = element_blank(),
    axis.text = element_text(size = axis_size),
    axis.ticks = element_line(linewidth = 0.2),
     axis.line = element_line(linewidth = 0, color = "black"),  # Set the axis line linewidth
   panel.border = element_rect(fill = NA, linewidth = 0.5),  # Set the panel border linewidth
  legend.key = element_rect(fill = "white", color = NA)  # Set legend key background to white and remove border
  )

print(pca)

cairo_pdf(paste0(Sys.Date(),"_pca_",exp, "_", treatment,".vs.", control,".pdf", sep = ""))
pca
dev.off()

## quartz_off_screen 
##                 2

11 Volcano plot

sequence <- c("treatment")

for (i in sequence) {
    set.seed(0)
    res <- results(dds_final, contrast = c("exp_9", i, "control"))

  gp2 <- wes_palettes$Zissou1
  key <- read.csv("../../raw-data/pa14_gene_key.csv",  header = TRUE)

  rownames(key) <- key$gene_id

  res <- merge(key, as.data.frame(res), by=0)
  res <- as.data.frame(res)
  res$Alias <- rownames(res)

 # Define gene list
  alg_genes <- c(
    "algA", "algF", "algJ", "algI", "algL", "algX", "algG", "algE",
    "algK", "alg44", "alg8", "algD", "mucC", "mucB", "algU", "algW",
    "algP", "algQ", "algR", "algZ", "algC", "algB"
  )

logfc_cutoff
keyvals <- ifelse(
 res$log2FoldChange <= -logfc_cutoff & res$padj <= pval_cutoff, gp2[1],
  ifelse(res$log2FoldChange >= logfc_cutoff & res$padj <= pval_cutoff, gp2[5],
         'lightgrey'))
keyvals[is.na(keyvals)] <- 'lightgrey'
names(keyvals)[keyvals == gp2[5]] <- 'Up-regulated'
names(keyvals)[keyvals == 'lightgrey'] <- 'NS'
names(keyvals)[keyvals == gp2[1]] <- 'Down-regulated'

volcano <- EnhancedVolcano(res,
                     lab = res[["name"]],
                     selectLab = alg_genes,
                     x = 'log2FoldChange',
                     y = 'padj',
                     xlab = bquote(~Log[2]~ 'fold change'),
                     pCutoff = pval_cutoff,
                     FCcutoff = logfc_cutoff,
                     cutoffLineType = "dashed",
                     pointSize = 1,
                     labSize = 3.0,
                     axisLabSize = 14,
                     labCol = 'black',
                     labFace = 'bold',
                     #col = c('grey', gp2[2], gp2[3], gp2[1]),
                     colCustom = keyvals,
                     colAlpha = 5/5,
                     legendPosition = 'bottom',
                     legendLabSize = 12,
                     legendIconSize = 4.0,
                     drawConnectors = TRUE,
                     arrowheads = FALSE,
                     widthConnectors = 0.2,
                     gridlines.major = FALSE,
                     borderWidth = 0.3,
                     gridlines.minor = FALSE,
                     title = paste0(treatment," vs ", control, sep = ""),
                     subtitle = "",
                     #caption = paste0("total = ", nrow(toptable), " variables"),
                     caption = paste0("Volcano plot with log2 FC of ", logfc_cutoff," and P value of ",pval_cutoff, "."),
                     max.overlaps = 30,
                     #ylim = c(0,10),
                     colConnectors = 'black') + 
      theme(axis.ticks=element_line(linewidth =0.2)) # +
 # scale_x_continuous(breaks = c(-5, -2.5, 0, 2.5, 5), limits = c(-6,6))

print(volcano)

cairo_pdf(paste0(Sys.Date(),"_volcano_", exp, "_", treatment,".vs.", control,".pdf", sep = ""))
print(volcano)
dev.off()
}

12 MA plot

sequence <- c("treatment")

for (i in sequence) {
    set.seed(0)

res <- results(dds_final, contrast = c("exp_9", i, "control"))

gp2 <- wes_palettes$Zissou1

res$gene_id <- rownames(res)
res <- as.data.frame(res)
head(res)
res <- full_join(key, as.data.frame(res), by="gene_id")
rownames(res) <- res$Alias

any(is.na(res))  # Check if there are any NAs

res <- res %>% filter(!is.na(baseMean))


ma <- ggmaplot(res, 
               fdr = pval_cutoff, 
               fc = 2^(logfc_cutoff), 
               size = 1,
               palette = c((gp2[5]), (gp2[1])),
               select.top.method = "fc",
             #  ylim = c(-15,15),
               legend = "bottom", 
               title = paste0(treatment," vs ", control, sep = ""),
              subtitle = paste0("Volcano plot with log2 FC of ", logfc_cutoff," and P value of ",pval_cutoff, "."),
               font.label = c( 12), 
             label.rectangle = FALSE,
              genenames = res$name,
              font.legend = c(12),
               font.main = c("bold", 14),
               alpha = 1.5,
               ggtheme = ggplot2::theme(panel.grid =element_blank(),
                                        axis.text = element_text(size = 12),
                                        axis.title = element_text(size = 14),
                                        panel.background = element_blank(),
                                        panel.border = element_rect(fill = NA))) 
print(ma)

cairo_pdf(paste0(Sys.Date(),"_ma_", exp, "_", treatment,".vs.", control,".pdf"))
print(ma)
dev.off()
}