library(mapa)3 Enrichment Analysis
The mapa package provides two powerful functions for pathway enrichment analysis: enrich_pathway() for Over-Representation Analysis (ORA) and do_gsea() for Gene Set Enrichment Analysis (GSEA). Both functions support multiple databases and can handle both gene and metabolite data (for metabolite, currently supports ORA only). Here we provide a unified, streamlined workflow that eliminates the need to run separate analyses for different databases, allowing you to analyze multiple databases simultaneously in a single function call.
Important
Prerequisites: Before running enrichment analysis, ensure your data has been properly preprocessed using the convert_id() function as described in Chapter 2 - Data Input and Preprocessing. The variable_info used in this chapter should be the output from the ID conversion step.
3.1 Gene-based Enrichment
3.1.1 Supported Databases and Keytypes
The mapa package leverages the powerful enrichment functions from the clusterProfiler package for pathway analysis. The following table shows which databases you can use for your organism and the supported key types:
| Database | Supported Organism | Keytype Options |
|---|---|---|
| Gene Ontology (GO) | Any organism with OrgDb object (organism annotation database package from Bioconductor or OrgDb object retrieved from AnnotationHub, see Section 2.2.1) | Any keyType supported by your OrgDb object |
| KEGG | All KEGG organisms | "kegg", "ncbi-geneid", "ncbi-proteinid", "uniprot" |
| Reactome | human, rat, mouse, celegans, yeast, zebrafish, fly, bovine, canine, chicken | ENTREZID only |
Tip
To check available GO keytype for your organism:
For model organisms with Bioconductor annotation packages
## For model organisms with Bioconductor annotation packages library(org.Hs.eg.db) AnnotationDbi::keytypes(org.Hs.eg.db) # [1] "ACCNUM" "ALIAS" "ENSEMBL" "ENSEMBLPROT" "ENSEMBLTRANS" "ENTREZID" "ENZYME" "EVIDENCE" "EVIDENCEALL" "GENENAME" # [11] "GENETYPE" "GO" "GOALL" "IPI" "MAP" "OMIM" "ONTOLOGY" "ONTOLOGYALL" "PATH" "PFAM" # [21] "PMID" "PROSITE" "REFSEQ" "SYMBOL" "UCSCKG" "UNIPROT"For non-model organisms with annotation OrgDb retrieved from AnnotationHub
# variable_info is the output from the ID conversion step in Chapter 2 - Data Input and Preprocessing AnnotationDbi::keytypes(variable_info$orgdb)For comprehensive understanding of enrichment analysis concepts, methodologies, we highly recommend reading the Biomedical Knowledge Mining using GOSemSim and clusterProfiler book. This book covers all the underlying methods that MAPA uses and will help you make informed decisions about your analysis parameters.
3.1.2 Basic Usage
Use enrich_pathway() for Over-Representation Analysis.
enriched_pathways <-
enrich_pathway(
variable_info = variable_info,
query_type = "gene",
database = c("go", "kegg", "reactome"),
# GO parameters
go.orgdb = org.Hs.eg.db,
go.keytype = "ENTREZID",
go.ont = "ALL",
# KEGG parameters
kegg.organism = "hsa",
kegg.keytype = "kegg",
# Reactome parameters
reactome.organism = "human",
# Statistical parameters
pvalueCutoff = 0.05,
pAdjustMethod = "BH"
)
# GO database...
# KEGG database...
# Reading KEGG annotation online: "https://rest.kegg.jp/link/hsa/pathway"...
# Reading KEGG annotation online: "https://rest.kegg.jp/list/pathway/hsa"...
# Reactome database...
# Done.The do_gsea() function performs GSEA using ranked gene lists based on fold changes or other metrics.
gsea_pathways <-
do_gsea(
variable_info = variable_info,
query_type = "gene",
order_by = "fc", # Column to rank genes by
database = c("go", "kegg", "reactome"),
# Database parameters
go.orgdb = org.Hs.eg.db,
go.ont = "ALL",
go.keytype = "ENTREZID",
kegg.organism = "hsa",
kegg.keytype = "kegg",
reactome.organism = "human",
# GSEA parameters
pvalueCutoff = 0.05,
pAdjustMethod = "BH"
)
Note
- Model organisms: use
variable_infodirectly and package name forgo.orgdb - Non-model organisms: use
variable_info$datafor parametervariable_infoandvariable_info$orgdbforgo.orgdb
3.2 Metabolite-based Enrichment
3.2.1 Supported Databases and Keytypes
| Database | Supported Organism | Keytype Options |
|---|---|---|
| KEGG | All KEGG organisms | KEGG compound IDs |
| SMPDB | Only for human | HMDB IDs |
3.2.2 Basic Usage
For metabolite data, ensure your variable_info contains HMDB IDs and/or KEGG compound IDs.
met_enriched_pathways <-
enrich_pathway(
variable_info = met_variable_info,
query_type = "metabolite",
database = c("hmdb", "metkegg"),
met_organism = "hsa",
save_to_local = TRUE,
pvalueCutoff = 0.05,
pAdjustMethod = "BH"
)3.3 Results interpretation
After enrichment analysis, view the enrichment analysis summary:
enriched_pathways
# --------------------
# Analysis method: enrich_pathway
# --------------------
# -----------Variable information------------
# 119 features/markers in total
# -----------Enrichment results and modules of genes------------
# -----------GO------------
# 1025 GO terms with p.adjust < 0.05
# No GO modules
# -----------KEGG------------
# 48 KEGG pathways with p.adjust < 0.05
# No KEGG modules
# -----------Reactome------------
# 48 Reactome pathways with p.adjust < 0.05
# No Reactome modules
# -----------Enrichment results and modules of metabolites------------
# -----------HMDB------------
# No HMDB results
# No HMDB modules
# -----------KEGG Metabolite------------
# No KEGG metabolite results
# No KEGG modules
# -----------Functional modules------------
# No Functional modules
# -----------LLM module interpretation------------
# No LLM module interpretation results
# --------------------
# Processing information
# 1 processings in total
# enrich_pathway ----------
# Package Function.used Time
# 1 mapa enrich_pathway() 2025-06-08 14:34:13You can access specific database results via @:
head(enriched_pathways@enrichment_go_result@result)
# ONTOLOGY ID Description GeneRatio BgRatio RichFactor FoldEnrichment zScore pvalue
# GO:0038084 BP GO:0038084 vascular endothelial growth factor signaling pathway 13/116 95/18805 0.13684211 22.18376 16.30756 2.147803e-14
# GO:0036005 BP GO:0036005 response to macrophage colony-stimulating factor 11/116 64/18805 0.17187500 27.86301 16.95936 1.713923e-13
# GO:0035924 BP GO:0035924 cellular response to vascular endothelial growth factor stimulus 13/116 122/18805 0.10655738 17.27424 14.20752 5.888098e-13
# GO:0006935 BP GO:0006935 chemotaxis 21/116 466/18805 0.04506438 7.30548 10.85886 8.851862e-13
# GO:0042330 BP GO:0042330 taxis 21/116 468/18805 0.04487179 7.27426 10.82885 9.608949e-13
# GO:0038145 BP GO:0038145 macrophage colony-stimulating factor signaling pathway 10/116 54/18805 0.18518519 30.02075 16.82498 1.024076e-12
# p_adjust qvalue geneID Count
# GO:0038084 5.859206e-11 3.635439e-11 3480/28514/1969/1956/64094/2050/5156/7422/8828/2064/5979/2324/780 13
# GO:0036005 2.337791e-10 1.450520e-10 3480/1435/1969/1956/6696/2050/5156/2064/5979/2324/780 11
# GO:0035924 4.656131e-10 2.888971e-10 3480/28514/1969/1956/64094/2050/5156/7422/8828/2064/5979/2324/780 13
# GO:0006935 4.656131e-10 2.888971e-10 6370/3958/5919/1435/7040/8633/56477/1969/58191/64094/5054/3491/5156/3569/6359/7422/3082/6360/10457/5328/8828 21
# GO:0042330 4.656131e-10 2.888971e-10 6370/3958/5919/1435/7040/8633/56477/1969/58191/64094/5054/3491/5156/3569/6359/7422/3082/6360/10457/5328/8828 21
# GO:0038145 4.656131e-10 2.888971e-10 3480/1435/1969/1956/2050/5156/2064/5979/2324/780 10head(gsea_pathways@enrichment_kegg_result@result)
# ID Description setSize enrichmentScore NES pvalue p_adjust qvalue rank
# hsa05014 hsa05014 Amyotrophic lateral sclerosis 11 0.6473595 2.500196 0.0001092759 0.0006556554 0.0003450818 218
# hsa05022 hsa05022 Pathways of neurodegeneration - multiple diseases 11 0.4649904 1.795860 0.0102663060 0.0307989180 0.0162099568 232
# hsa05010 hsa05010 Alzheimer disease 10 0.4709263 1.667865 0.0220601399 0.0441202798 0.0232211999 214
# leading_edge core_enrichment Count
# hsa05014 tags=100%, list=36%, signal=65% 6390/23435/4720/5690/56893/6391/55706/842/5710/22926/55746 11
# hsa05022 tags=91%, list=39%, signal=57% 6390/23435/4720/5690/6391/1435/842/5710/22926/9927 10
# hsa05010 tags=90%, list=36%, signal=59% 6390/10000/4720/5690/6391/1435/842/5710/22926 9Key columns in enrichment results:
Core Information:
ID: Pathway identifier (e.g., GO:0042060 for GO, hsa04060 for KEGG, R-HSA-5669034 for Reactome)Description: Pathway nameONTOLOGY: Biological ontology (for GO only: MF=molecular function, CC=cellular component, BP=biological process)category/subcategory: KEGG pathway categories (for KEGG only)
Statistical Measures:
pvalue: Raw p-value from hypergeometric test (equivalent to one-sided Fisher’s exact test)p_adjust: Adjusted p-value after multiple testing correction (BH method by default)qvalue: Q-value for FDR control. For more information, see?qvalue::qvalue
Gene Mapping:
GeneRatio: Ratio of input genes annotated to this pathway (format: “genes_in_pathway/total_input_genes”)BgRatio: Ratio of all genes annotated to this pathway in the background universe (format: “pathway_genes/universe_genes”)Count: Total number of genes from input list that match this pathwaygeneID: Gene IDs that overlap between your gene list and the pathway (separated by “/”)
Enrichment Metrics:
RichFactor: Ratio of input genes annotated to a pathway vs. all genes annotated to this pathwayFoldEnrichment: Enrichment fold change (GeneRatio divided by BgRatio)zScore: Standard deviations away from expected overlap (How unusual or extreme the observed enrichment is compared to what you’d expect by chance. Higher absolute z-score means more “surprising” or significant enrichment.)
For GSEA results, additional columns specific to ranked list analysis include:
setSize: Total number of genes in the gene set/pathwayenrichmentScore: Degree of overrepresentation at top/bottom of ranked listNES: Normalized Enrichment Score - main metric for interpretation- Positive NES: pathway enriched in upregulated genes (pathway activation)
- Negative NES: pathway enriched in downregulated genes (pathway suppression)
rank: Position in ranked list where maximum enrichment score occurredleading_edge: Statistics for the leading-edge subset of genescore_enrichment: Core enriched genes that contribute most to the enrichment signalCount: Number of core enriched genes
For metabolite ORA results, the result structure differs from gene-based analysis:
Pathway Information:
pathway_id: Small Molecule Pathway Database Pathway identifier (e.g., SMP0000028)pathway_name: Name of the metabolic pathway (e.g., “Caffeine Metabolism”)describtion: Detailed description of the pathway from the corresponding databasepathway_class: Classification of pathway (e.g., “Metabolic;primary_pathway”, “Disease;primary_pathway”)
Statistical Measures:
p_value: Raw p-value from enrichment testp_adjust: Adjusted p-value for multiple testing correction
Metabolite Mapping:
all_id: All metabolite IDs in this pathway (separated by “;”)all_number: Total number of metabolites in the pathwaymapped_id: Your input metabolite IDs that map to this pathway (separated by “;”)mapped_number: Number of your metabolites that map to this pathwaymapped_percentage: Percentage of pathway metabolites covered by your input data
The enrichment results provide the foundation for downstream similarity analysis and functional module identification in the MAPA workflow.