1. The Complete End-to-End Plant RNA-Seq Pipeline

Scientific reproducibility demands that every analytical phase—from raw reads to final publication figures—be executable via automated scripts. Below is the master Quarto/RMarkdown workflow template for the Solanum tuberosum drought transcriptomics investigation.

2. Master R Script for Automated Execution

###############################################################################
# Plant Stress RNA-Seq Analysis Workflow: Solanum tuberosum
# Author: OMICS Portal Curriculum Core
# License: MIT Open Access
###############################################################################

suppressPackageStartupMessages({
  library(tidyverse)
  library(DESeq2)
  library(apeglm)
  library(pheatmap)
  library(EnhancedVolcano)
  library(clusterProfiler)
  library(openxlsx)
})

# 1. Load Experimental Metadata and Count Matrix
targets <- read.csv("sample_metadata.csv", stringsAsFactors = FALSE)
counts_raw <- read.table("potato_gene_counts.txt", header = TRUE, row.names = 1, skip = 1)
counts <- as.matrix(counts_raw[, 6:ncol(counts_raw)])
colnames(counts) <- targets$Sample_ID

# 2. Build DESeq2 Object and Pre-Filter Low Counts (>= 10 reads in at least 3 samples)
dds <- DESeqDataSetFromMatrix(
  countData = counts,
  colData   = targets,
  design    = ~ Genotype * Drought_Stage
)
keep <- rowSums(counts(dds) >= 10) >= 3
dds <- dds[keep, ]

# 3. Model Fitting
dds <- DESeq(dds)

# 4. Extract Contrasts & Moderation
res_d7 <- results(dds, contrast = c("Drought_Stage", "D7_Severe", "D0_Control"), alpha = 0.05)
res_d7_shrunk <- lfcShrink(dds, coef = "Drought_Stage_D7_Severe_vs_D0_Control", type = "apeglm")

# 5. Export Publication Results Table
res_export <- as.data.frame(res_d7_shrunk) %>%
  rownames_to_column(var = "Gene_ID") %>%
  arrange(padj)

write.xlsx(res_export, file = "Table_S1_Drought_DEGs_Solanum_tuberosum.xlsx", asTable = TRUE)
cat("Complete end-to-end plant transcriptomics pipeline executed successfully!\n")