ReactomePA
Reactome is a free, open-source, curated and peer-reviewed pathway database. Their goal is to provide intuitive bioinformatics tools for the visualization, interpretation and analysis of pathway knowledge.
Dependencies & Environment
Package-intrinsic requirements from the Bioconductor landing page — reproduce in any R environment.
- Version: 1.56.0 · Bioconductor: 3.23 · R: ≥ 4.6
- Imports: AnnotationDbi, enrichplot, enrichit, ggplot2, ggraph, reactome.db, igraph, graphite, gson, yulab.utils
- Install:
BiocManager::install("ReactomePA")
When to Use
- Identifying significantly enriched Reactome pathways from differentially expressed gene lists
- Functional interpretation of RNA-seq or microarray results
- Comparative pathway analysis across supported model organisms
- Studies requiring curated, peer-reviewed pathway annotations
When NOT to Use
- Pathway analysis requiring detailed network topology or edge information
- Studies needing pathway visualization beyond enrichment statistics
- Organisms outside the seven supported species
- Gene lists without Entrez ID mapping
- Analyses requiring alternative pathway databases (KEGG, Biocarta, etc.)
Scientific Assumptions
- {'assumption': 'Input gene list represents true biological signal (differentially expressed genes) rather than technical artifacts', 'violation_context': 'When DE analysis used inappropriate statistical methods, insufficient replicates, or poor quality control'}
- {'assumption': 'Reactome pathway annotations are complete and accurate for the organism being studied', 'violation_context': 'Newly discovered pathways or organism-specific variations not yet curated in Reactome'}
- {'assumption': 'Gene set enrichment follows hypergeometric distribution (genes are randomly distributed across pathways)', 'violation_context': 'When genes have strong co-expression patterns or functional clustering independent of pathway membership'}
- {'assumption': 'Multiple testing correction appropriately controls error rate for the number of pathways tested', 'violation_context': 'When pathway annotations are highly correlated or overlapping, standard corrections may be conservative'}
- {'assumption': 'Entrez ID mapping is accurate and one-to-one between input genes and Reactome annotations', 'violation_context': 'When gene ID conversions are outdated or when one gene maps to multiple Entrez IDs'}
Common Pitfalls
- {'mistake': 'Using gene symbols or other identifiers instead of Entrez IDs', 'consequence': 'Tool will fail to match genes to Reactome pathways, resulting in no enrichment results or errors', 'recommendation': "Convert gene identifiers to Entrez IDs before input. Use an ID-mapping tool such as AnnotationDbi or biomaRt"}
- {'mistake': 'Analyzing organisms not in the supported list', 'consequence': 'Tool will not run or produce incorrect results due to missing organism-specific pathway annotations', 'recommendation': 'Verify organism is in supported list: human, rat, mouse, C. elegans, yeast, zebrafish, or fly'}
- {'mistake': 'Not adjusting p-value threshold for multiple testing correction', 'consequence': 'Inflated false positive rate when reporting enriched pathways', 'recommendation': 'Use appropriate multiple testing correction method (default BH/Benjamini-Hochberg recommended for FDR control)'}
- {'mistake': 'Including very small or very large gene sets without filtering', 'consequence': 'Spurious enrichment from gene sets with few genes or loss of specificity from very large sets', 'recommendation': 'Use default gene set size filtering (10-500 genes) or adjust based on biological context'}
Key Parameters
- {'parameter': 'P-value threshold', 'scientific_meaning': 'Cutoff for determining statistical significance of pathway enrichment. Controls false positive rate.', 'typical_values': '0.05 (default), 0.01 for stringent analysis, 0.1 for exploratory analysis', 'context_guidance': {'large_gene_lists': '0.05 or more stringent', 'small_gene_lists': '0.05-0.1', 'exploratory_analysis': '0.1', 'publication_quality': '0.05 with multiple testing correction'}, 'quote': 'This is the cutoff p-value to determine significant enrichment of a pathway. Default is 0.05.'}
- {'parameter': 'Method to adjust pathway p-values for multiple testing', 'scientific_meaning': 'Statistical correction for multiple hypothesis testing to control family-wise error rate or false discovery rate', 'typical_values': 'BH (Benjamini-Hochberg), holm, hochberg, hommel, bonferroni, BY, fdr, none', 'context_guidance': {'recommended': 'BH (default) - controls FDR, good balance between sensitivity and specificity', 'conservative': 'bonferroni - controls family-wise error rate, most stringent', 'exploratory': 'none - no correction, highest sensitivity but highest false positive rate'}, 'quote': "Choose one method to adjust pathway p-values. Included methods are: 'BH', 'holm', 'hochberg', 'hommel', 'bonferroni', 'BY', 'fdr' and 'none'. Default is 'BH'."}
- {'parameter': 'Minimum and maximum size of each geneSet', 'scientific_meaning': 'Filters gene sets by size to avoid spurious enrichment from very small sets or loss of specificity from very large sets', 'typical_values': 'Minimum: 5-15 genes, Maximum: 300-500 genes (default 10-500)', 'context_guidance': {'default': '10-500 genes - balanced approach', 'stringent': '15-300 genes - removes very small and very large sets', 'exploratory': '5-1000 genes - more permissive'}, 'quote': 'Set minimum and maximum size of each gene set to be included in the analysis. Default are 10 and 500.'}
- {'parameter': 'Convert gene ID to SYMBOL', 'scientific_meaning': 'Option to convert Entrez IDs to human-readable gene symbols in output for interpretation', 'typical_values': 'Yes or No (default: No)', 'context_guidance': {'publication': 'Yes - improves readability for figures and tables', 'downstream_analysis': 'No - maintain Entrez IDs for computational pipelines', 'interpretation': 'Yes - easier to identify genes of interest'}, 'quote': 'If you want to convert entrez gene ID to gene SYMBOL name, choose Yes. Default: No.'}
Result Interpretation
- {'guidance': 'Output CSV contains enriched pathways with columns: ID (Reactome pathway ID), Description (pathway name), GeneRatio (number of input genes in pathway / total input genes), BgRatio (pathway size / total Reactome genes). Higher GeneRatio indicates stronger enrichment. Pathways are ranked by adjusted p-value.', 'quote': 'A csv file including all enriched pathways. Example: ID, Description, GeneRatio, BgRatio'}
- {'guidance': 'GeneRatio represents the proportion of your input genes that map to each pathway. A GeneRatio of 22/326 means 22 of your 326 input genes are in that pathway. Higher ratios indicate more of your genes are involved in that pathway.'}
- {'guidance': 'BgRatio represents the background pathway size relative to total Reactome genes. This helps assess whether enrichment is due to pathway size bias. Compare GeneRatio to BgRatio to evaluate enrichment strength.'}
- {'guidance': 'Validate enriched pathways by: (1) checking biological relevance to experimental conditions, (2) examining which genes drive enrichment, (3) comparing to literature on pathway involvement in your biological system'}
Alternatives
- {'tool': 'clusterProfiler', 'when_to_prefer_this': 'ReactomePA when specifically using Reactome pathway database and working with supported organisms', 'when_to_prefer_alternative': 'clusterProfiler when needing multiple pathway databases (KEGG, GO, Reactome) or more flexible organism support'}
- {'tool': 'KEGG pathway analysis tools', 'when_to_prefer_this': 'ReactomePA when requiring curated, peer-reviewed Reactome pathways with detailed mechanistic annotations', 'when_to_prefer_alternative': 'KEGG tools when needing broader organism coverage or KEGG-specific pathway knowledge'}
- {'tool': 'Enrichr', 'when_to_prefer_this': 'ReactomePA when requiring programmatic analysis and pipeline integration', 'when_to_prefer_alternative': 'Enrichr when needing interactive web interface and multiple library options'}
Citations
- Yu G, He QY. (2016). ReactomePA: an R/Bioconductor package for reactome pathway analysis and visualization. Molecular BioSystems, 12(2):477-479. PMID:26661513 doi:10.1039/C5MB00663E
References
Run this on BioMate
This skill is the knowledge layer — when, why, and how to use reactomepa. To run this analysis on your own data with managed compute, automated QC, and reproducible outputs, use BioMate — free to start.
