bio-workflows-expression-to-pathways

name	bio-workflows-expression-to-pathways
description	Workflow from differential expression results to functional enrichment analysis. Covers GO, KEGG, Reactome enrichment with clusterProfiler and visualization. Use when taking DE results to pathway enrichment.
tool_type	r
primary_tool	clusterProfiler
workflow	true
depends_on	["pathway-analysis/go-enrichment","pathway-analysis/kegg-pathways","pathway-analysis/reactome-pathways","pathway-analysis/gsea","pathway-analysis/enrichment-visualization"]
qc_checkpoints	[{"input_validation":"Valid gene IDs, sufficient DE genes"},{"enrichment_qc":"Reasonable number of terms, p-values not all significant"}]

Expression to Pathways Workflow

Convert differential expression results into biological insights through functional enrichment analysis.

Workflow Overview

DE Results (gene list or ranked list)
    |
    v
[1. Gene ID Conversion] --> Convert to Entrez/Ensembl
    |
    v
[2. Over-representation Analysis]
    |
    +---> GO Enrichment (BP, MF, CC)
    |
    +---> KEGG Pathways
    |
    +---> Reactome Pathways
    |
    v
[3. GSEA (ranked genes)]
    |
    v
[4. Visualization] -----> Dot plots, networks, bar plots
    |
    v
Functional annotations and pathway insights

Input Preparation

From DESeq2 Results

library(DESeq2)
library(clusterProfiler)
library(org.Hs.eg.db)

# Load DE results
res <- read.csv('deseq2_results.csv', row.names = 1)

# Significant genes for ORA
sig_genes <- rownames(subset(res, padj < 0.05 & abs(log2FoldChange) > 1))

# All genes for background
all_genes <- rownames(res)

# Ranked list for GSEA (by stat or log2FC)
ranked_genes <- res$log2FoldChange
names(ranked_genes) <- rownames(res)
ranked_genes <- sort(ranked_genes, decreasing = TRUE)
ranked_genes <- ranked_genes[!is.na(ranked_genes

Gene ID Conversion

# Convert gene symbols to Entrez IDs
sig_entrez <- bitr(sig_genes, fromType = 'SYMBOL', toType = 'ENTREZID',
                   OrgDb = org.Hs.eg.db)

# For ranked list
ranked_entrez <- bitr(names(ranked_genes), fromType = 'SYMBOL', toType = 'ENTREZID',
                      OrgDb = org.Hs.eg.db)
ranked_list <- ranked_genes[ranked_entrez$SYMBOL]
names(ranked_list) <- ranked_entrez$ENTREZID

Step 1: GO Over-representation Analysis

# Biological Process
go_bp <- enrichGO(gene = sig_entrez$ENTREZID,
                  OrgDb = org.Hs.eg.db,
                  ont = 'BP',
                  pAdjustMethod = 'BH',
                  pvalueCutoff = 0.05,
                  qvalueCutoff = 0.1,
                  readable = TRUE)

# Molecular Function
go_mf <- enrichGO(gene = sig_entrez$ENTREZID,
                  OrgDb = org.Hs.eg.db,
                  ont = 'MF',
                  pAdjustMethod = 'BH',
                  pvalueCutoff = 0.05,
                  readable = TRUE)

# Cellular Component
go_cc <- enrichGO(gene = sig_entrez$ENTREZID,
                  OrgDb = org.Hs.eg.db,
                  ont  
                  pAdjustMethod  
                  pvalueCutoff  
                  readable  


go_bp_simple  simplifygo_bp cutoff   by  

Step 2: KEGG Pathway Enrichment

kegg <- enrichKEGG(gene = sig_entrez$ENTREZID,
                   organism = 'hsa',
                   pvalueCutoff = 0.05,
                   qvalueCutoff = 0.1)

# Convert KEGG IDs to readable names
kegg <- setReadable(kegg, OrgDb = org.Hs.eg.db, keyType = 'ENTREZID')

Step 3: Reactome Pathway Enrichment

library(ReactomePA)

reactome <- enrichPathway(gene = sig_entrez$ENTREZID,
                          organism = 'human',
                          pvalueCutoff = 0.05,
                          readable = TRUE)

Step 4: Gene Set Enrichment Analysis (GSEA)

# GO GSEA
gsea_go <- gseGO(geneList = ranked_list,
                 OrgDb = org.Hs.eg.db,
                 ont = 'BP',
                 minGSSize = 10,
                 maxGSSize = 500,
                 pvalueCutoff = 0.05,
                 verbose = FALSE)

# KEGG GSEA
gsea_kegg <- gseKEGG(geneList = ranked_list,
                     organism = 'hsa',
                     minGSSize = 10,
                     maxGSSize = 500,
                     pvalueCutoff = 0.05,
                     verbose = FALSE)

Step 5: Visualization

library(enrichplot)
library(ggplot2)

# Dot plot
dotplot(go_bp_simple, showCategory = 20) +
    ggtitle('GO Biological Process Enrichment')
ggsave('go_bp_dotplot.pdf', width = 10, height = 8)

# Bar plot
barplot(kegg, showCategory = 15) +
    ggtitle('KEGG Pathway Enrichment')
ggsave('kegg_barplot.pdf', width = 9, height = 6)

# Enrichment map (network of related terms)
go_bp_simple <- pairwise_termsim(go_bp_simple)
emapplot(go_bp_simple, showCategory = 30) +
    ggtitle('GO Term Similarity Network'
ggsave width   height  


cnetplotgo_bp showCategory   categorySize   
    ggtitle
ggsave width   height  


gseaplot2gsea_kegg geneSetID   pvalue_table  
ggsave width   height  


ridgeplotgsea_go showCategory  
ggsave width   height  

Step 6: Export Results

# Export enrichment results
write.csv(as.data.frame(go_bp), 'go_bp_enrichment.csv', row.names = FALSE)
write.csv(as.data.frame(kegg), 'kegg_enrichment.csv', row.names = FALSE)
write.csv(as.data.frame(reactome), 'reactome_enrichment.csv', row.names = FALSE)
write.csv(as.data.frame(gsea_go), 'gsea_go_results.csv', row.names = FALSE)

# Combine key results
combined <- rbind(
    data.frame(Database = 'GO_BP', as.data.frame(go_bp_simple)[1:10,]),
    data.frame(Database   as.data.framekegg
    data.frameDatabase   as.data.framereactome

write.csvcombined  row.names  

Parameter Recommendations

Analysis	Parameter	Value
enrichGO	pvalueCutoff	0.05
enrichGO	qvalueCutoff	0.1
simplify	cutoff	0.7
gseGO	minGSSize	10
gseGO	maxGSSize	500
GSEA	perm	1000 (default)

Troubleshooting

Issue	Likely Cause	Solution
No enriched terms	Too few genes, wrong IDs	Check gene IDs, relax thresholds
All terms significant	Too many genes	Be more stringent with DE cutoffs
Gene ID conversion fails	Wrong organism, format	Check OrgDb package, gene format
GSEA no results	Poor ranking, small gene sets	Check ranked list, adjust minGSSize

Complete Workflow Script

library(clusterProfiler)
library(org.Hs.eg.db)
library(ReactomePA)
library(enrichplot)
library(ggplot2)

# Configuration
de_file <- 'deseq2_results.csv'
output_dir <- 'pathway_analysis'
dir.create(output_dir, showWarnings = FALSE)

# Load and prepare data
res <- read.csv(de_file, row.names = 1)
sig_genes <- rownames(subset(res, padj < 0.05 & abs(log2FoldChange) > 1))
cat('Significant genes:', length(sig_genes), '\n')

# Convert IDs
sig_entrez <- bitr(sig_genes, fromType   toType   OrgDb  org.Hs.eg.db
cat nrowsig_entrez 


ranked  reslog2FoldChange
ranked  rownamesres
ranked  sortrankedranked decreasing  
ranked_entrez  bitrranked fromType   toType   OrgDb  org.Hs.eg.db
ranked_list  rankedranked_entrezSYMBOL
ranked_list  ranked_entrezENTREZID


go_bp  enrichGOsig_entrezENTREZID OrgDb  org.Hs.eg.db ont   readable  
go_bp_simple  simplifygo_bp cutoff  


kegg  enrichKEGGsig_entrezENTREZID organism  
kegg  setReadablekegg OrgDb  org.Hs.eg.db keyType  


reactome  enrichPathwaysig_entrezENTREZID organism   readable  


gsea_go  gseGOranked_list OrgDb  org.Hs.eg.db ont   verbose  


pdffile.pathoutput_dir  width   height  
printdotplotgo_bp_simple showCategory    ggtitle
printbarplotkegg showCategory    ggtitle
 nrowas.data.framereactome   
    printdotplotreactome showCategory    ggtitle

dev.off


write.csvas.data.framego_bp_simple file.pathoutput_dir  row.names  
write.csvas.data.framekegg file.pathoutput_dir  row.names  
write.csvas.data.framereactome file.pathoutput_dir  row.names  

cat output_dir 
cat nrowas.data.framego_bp_simple 
cat nrowas.data.framekegg 
cat nrowas.data.framereactome 

Related Skills

• pathway-analysis/go-enrichment - GO enrichment details
• pathway-analysis/kegg-pathways - KEGG analysis
• pathway-analysis/reactome-pathways - Reactome analysis
• pathway-analysis/gsea - GSEA methods
• pathway-analysis/enrichment-visualization - Visualization options