| name | bulkrna-de |
| description | Differential expression analysis via PyDESeq2 with Welch's t-test fallback — volcano plots, MA plots, p-value diagnostics. |
| version | 0.3.0 |
| author | OmicsClaw |
| license | MIT |
| tags | ["bulkrna","differential-expression","DESeq2","volcano","MA-plot","fold-change"] |
| requires | ["numpy","pandas","matplotlib","scipy"] |
| metadata | {"omicsclaw":{"domain":"bulkrna","emoji":"🔬","trigger_keywords":["differential expression","DE analysis","DESeq2","volcano plot","fold change","DEGs","bulk DE"],"allowed_extra_flags":["--control-prefix","--lfc-cutoff","--method","--padj-cutoff","--treat-prefix"],"legacy_aliases":["bulk-de"],"saves_h5ad":false}} |
Bulk RNA-seq Differential Expression
Differential expression analysis for bulk RNA-seq count matrices. Primary engine is PyDESeq2 (Python implementation of the DESeq2 model); falls back to Welch's t-test with log2FC and Benjamini-Hochberg FDR correction when PyDESeq2 is not installed.
Core Capabilities
- Full DESeq2-style negative binomial GLM via PyDESeq2, or Welch's t-test fallback
- Produces columns consistent with DESeq2 output: gene, baseMean, log2FoldChange, lfcSE, stat, pvalue, padj
- Volcano plot with top gene labels and MA plot with threshold lines
- P-value distribution histogram for diagnostic assessment
- Automatic sample group detection from column name prefixes
- Benjamini-Hochberg FDR correction with proper NaN handling
CLI Reference
python omicsclaw.py run bulkrna-de --demo
python omicsclaw.py run bulkrna-de --input <counts.csv> --output <dir>
python bulkrna_de.py --input counts.csv --output results/
python bulkrna_de.py --demo --output /tmp/de_demo
python bulkrna_de.py --input counts.csv --output results/ --method ttest --ctrl-prefix ctrl --treat-prefix treat
python bulkrna_de.py --input counts.csv --output results/ --padj-cutoff 0.01 --lfc-cutoff 1.5
Why This Exists
- Without it: Researchers must install R and DESeq2, write design formulas, handle normalization, shrinkage, and multiple testing correction manually, then create separate visualization scripts.
- With it: A single Python command runs the full DE pipeline — from raw counts to filtered DEG tables, volcano plots, MA plots, and p-value diagnostic histograms.
- Why OmicsClaw: Provides a pure-Python DE workflow via PyDESeq2 with graceful fallback to t-test, integrated into the OmicsClaw reporting framework.
Algorithm / Methodology
PyDESeq2 (Primary)
- Construct
DeseqDataSet from count matrix and sample metadata
- Estimate size factors (median-of-ratios normalization)
- Estimate dispersions (Cox-Reid profile-adjusted maximum likelihood)
- Fit negative binomial GLM and perform Wald test
- Apply Benjamini-Hochberg FDR correction
Welch's t-test (Fallback)
- Compute log2 fold change:
log2(mean_treat + 1) - log2(mean_ctrl + 1)
- Estimate standard error via delta method approximation
- Welch's t-test (unequal variance) per gene
- Benjamini-Hochberg FDR with NaN-safe handling
Key Parameters
