Multi-Omics Analysis of Multiple Sclerosis

Task 7: Functional Enrichment Analysis of DEGs and DMR-Associated Genes

This task performs Gene Ontology (GO) enrichment analysis to explore the biological significance of differentially expressed genes (DEGs) and genes associated with differentially methylated regions (DMRs). The workflow includes:

• Loading filtered DEG and DMR datasets
• Annotating CpG sites and mapping to gene symbols
• Converting gene symbols to Entrez IDs
• Running GO enrichment analysis (Biological Process category)
• Visualizing enriched terms using dot plots

The analysis highlights key biological processes potentially regulated by transcriptional and epigenetic alterations in the studied condition.

---

Load All Required Libraries for Multi-Omics Analysis

Load DEG and DMR Data + Extract Gene Symbols

# Load DEG table (make sure rownames are ENTREZID if used directly)
deg_filtered <- read.csv("DEG_significant_filtered.csv", row.names = 1)

# Load DMR data
dmrs_lm <- read.csv("dmrs_methylation_lm_based.csv", row.names = 1)

# Get CpG annotation from 450k array
ann450k <- getAnnotation(IlluminaHumanMethylation450kanno.ilmn12.hg19)

# Map CpG IDs to gene symbols
mapped_anns <- ann450k[rownames(ann450k) %in% rownames(dmrs_lm), ]
dmr_genes <- unique(na.omit(mapped_anns$UCSC_RefGene_Name))
dmr_genes <- unique(unlist(strsplit(dmr_genes, ";")))  # handle multiple gene symbols

Convert Gene Symbols to Entrez IDs

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

## 'select()' returned 1:1 mapping between keys and columns

## Warning in bitr(dmr_genes, fromType = "SYMBOL", toType = "ENTREZID", OrgDb =
## org.Hs.eg.db): 11.96% of input gene IDs are fail to map...

Start of Enrichment Analysis

#Enrichment Analysis
ego <- enrichGO(gene          = rownames(deg_filtered),
                OrgDb         = org.Hs.eg.db,
                keyType       = "ENTREZID",  # Important
                ont           = "BP",
                pAdjustMethod = "BH",
                pvalueCutoff  = 0.05,
                readable      = TRUE)

# View results
dotplot(ego)

# Install annotation package if not already
# BiocManager::install("IlluminaHumanMethylation450kanno.ilmn12.hg19")
library(IlluminaHumanMethylation450kanno.ilmn12.hg19)
library(minfi)  # needed for getAnnotation

# Get the CpG annotation
ann450k <- getAnnotation(IlluminaHumanMethylation450kanno.ilmn12.hg19)

dmrs <- dmrs_lm

# Match CpG IDs in annotation
mapped_anns <- ann450k[rownames(ann450k) %in% rownames(dmrs), ]

# Extract gene symbols associated with DMRs
dmr_genes <- unique(na.omit(mapped_anns$UCSC_RefGene_Name))
dmr_genes <- unique(unlist(strsplit(dmr_genes, ";")))  # handle multiple genes per CpG

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

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

## 'select()' returned 1:1 mapping between keys and columns

## Warning in bitr(dmr_genes, fromType = "SYMBOL", toType = "ENTREZID", OrgDb =
## org.Hs.eg.db): 11.96% of input gene IDs are fail to map...

# Run enrichment analysis (Gene Ontology - Biological Process)
ego <- enrichGO(
  gene = entrez_ids_dmr$ENTREZID,
  OrgDb = org.Hs.eg.db,
  ont = "BP",             # BP = Biological Process
  pAdjustMethod = "BH",
  pvalueCutoff = 0.05,
  readable = TRUE
)

# Plot the enrichment result
dotplot(ego, showCategory = 20, title = "GO Biological Process Enrichment")