bio-differential-expression-batch-correction
Remove batch effects from RNA-seq data using ComBat, ComBat-Seq, limma removeBatchEffect, and SVA for unknown batch variables
Install / Use
npx skills add FreedomIntelligence/OpenClaw-Medical-Skills --skill bio-differential-expression-batch-correctionInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Tags
Our assessment of bio-differential-expression-batch-correction
bio-differential-expression-batch-correction scores 89/100 on our quality scale, 1581st of 4,700 Development & Engineering skills we index (top 34%).
Its SKILL.md is 8.0 KB long, well organised into 53 sections with 10 code examples: a thorough specification that gives an agent plenty to work with.
With 3,034 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
- Our last check on 2026-10-31 found the source still online.
- No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
- Its trust signals score 86/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-09-29. Automated pattern scan on 2026-09-29. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
bio-differential-expression-batch-correction compared with similar skills
All 4 of these similar skills score higher than bio-differential-expression-batch-correction; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| bio-differential-expression-batch-correction (this skill)by FreedomIntelligence | 89 | 3.0k | 4mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 116.8k | 8d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.1k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.3k | 2d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.6k | today | MCP Server |
Frequently asked questions
- How do I install bio-differential-expression-batch-correction?
- Run
npx skills add FreedomIntelligence/OpenClaw-Medical-Skills --skill bio-differential-expression-batch-correction. The install tabs above show the steps for each supported agent. - Which AI agents does bio-differential-expression-batch-correction work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is bio-differential-expression-batch-correction safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 86/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is bio-differential-expression-batch-correction still maintained?
- The repository was last updated about 4 months ago. That is recent enough to be usable, but agent tooling moves fast, so check the instructions against your agent's current version.
Skill content
View source on GitHubname: bio-differential-expression-batch-correction description: Remove batch effects from RNA-seq data using ComBat, ComBat-Seq, limma removeBatchEffect, and SVA for unknown batch variables. Use when correcting batch effects in expression data. tool_type: r primary_tool: sva
Version Compatibility
Reference examples tested with: DESeq2 1.42+, ggplot2 3.5+, limma 3.58+, scanpy 1.10+
Before using code patterns, verify installed versions match. If versions differ:
- R:
packageVersion('<pkg>')then?function_nameto verify parameters
If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.
Batch Effect Correction
ComBat-Seq (Count Data)
Goal: Remove batch effects from raw count data while preserving biological group differences.
Approach: Apply ComBat-Seq's negative binomial regression to adjust counts, keeping the integer nature of the data.
"Remove batch effects from my RNA-seq counts" → Adjust raw count matrix for known batch labels using negative binomial modeling, preserving biological condition effects.
library(sva)
# counts: raw count matrix (genes x samples)
# batch: vector of batch labels
# group: vector of biological condition (optional, to preserve)
corrected_counts <- ComBat_seq(counts = as.matrix(counts),
batch = batch,
group = condition,
full_mod = TRUE)
# Result is batch-corrected count matrix
# Use for visualization, clustering, but NOT for DE (use design formula instead)
ComBat (Normalized Data)
Goal: Remove batch effects from normalized (log-transformed or TPM) expression data.
Approach: Apply parametric empirical Bayes adjustment to normalized expression while protecting biological covariates.
library(sva)
# For normalized expression (log-transformed, TPM, etc.)
# NOT for raw counts
# Create model matrix
mod <- model.matrix(~ condition, data = metadata)
mod0 <- model.matrix(~ 1, data = metadata)
# Run ComBat
corrected_expr <- ComBat(dat = as.matrix(normalized_expr),
batch = metadata$batch,
mod = mod,
par.prior = TRUE)
limma removeBatchEffect
Goal: Produce batch-corrected expression values for visualization while preserving group differences.
Approach: Regress out the batch effect from normalized expression using limma's linear model.
library(limma)
# For visualization/clustering only
# Preserves group differences while removing batch
design <- model.matrix(~ condition, data = metadata)
corrected_expr <- removeBatchEffect(normalized_expr,
batch = metadata$batch,
design = design)
# For PCA, heatmaps, etc.
DESeq2 Design Formula (Recommended for DE)
Goal: Account for batch effects during DE testing without modifying the count data.
Approach: Include batch as a covariate in the DESeq2 design formula so batch variance is modeled, not removed.
library(DESeq2)
# Include batch in design formula - preferred for DE analysis
dds <- DESeqDataSetFromMatrix(countData = counts,
colData = metadata,
design = ~ batch + condition)
# Batch is modeled, not removed
# DE results are adjusted for batch
dds <- DESeq(dds)
res <- results(dds, contrast = c('condition', 'treatment', 'control'))
Surrogate Variable Analysis (SVA)
Goal: Discover and correct for unknown sources of variation (hidden batch effects).
Approach: Estimate surrogate variables from the residual variation not explained by the biological model.
"Correct for unknown batch effects in my expression data" → Estimate latent surrogate variables capturing unwanted variation, then include them as covariates in the DE model.
library(sva)
# When batch is unknown, estimate surrogate variables
mod <- model.matrix(~ condition, data = metadata)
mod0 <- model.matrix(~ 1, data = metadata)
# Estimate number of surrogate variables
n_sv <- num.sv(normalized_expr, mod, method = 'leek')
# Estimate surrogate variables
svobj <- sva(normalized_expr, mod, mod0, n.sv = n_sv)
# Add SVs to design for DE
design_with_sv <- cbind(mod, svobj$sv)
SVA with DESeq2
Goal: Integrate surrogate variables into DESeq2 to adjust for hidden confounders during DE testing.
Approach: Estimate SVs from normalized counts, add them to colData, and update the design formula.
library(DESeq2)
library(sva)
# Normalize for SV estimation
dds <- DESeqDataSetFromMatrix(countData = counts, colData = metadata, design = ~ condition)
dds <- estimateSizeFactors(dds)
norm_counts <- counts(dds, normalized = TRUE)
# Estimate SVs
mod <- model.matrix(~ condition, data = metadata)
mod0 <- model.matrix(~ 1, data = metadata)
svobj <- sva(norm_counts, mod, mod0)
# Add SVs to colData
for (i in seq_len(ncol(svobj$sv))) {
colData(dds)[[paste0('SV', i)]] <- svobj$sv[, i]
}
# Update design
sv_formula <- as.formula(paste('~', paste(paste0('SV', 1:ncol(svobj$sv)), collapse = ' + '), '+ condition'))
design(dds) <- sv_formula
# Run DESeq2
dds <- DESeq(dds)
Visualize Batch Effects
Goal: Confirm batch effect removal by comparing PCA plots before and after correction.
Approach: Run PCA on pre- and post-correction expression, coloring points by batch and condition.
library(ggplot2)
# PCA before correction
pca_before <- prcomp(t(normalized_expr), scale. = TRUE)
pca_df <- data.frame(PC1 = pca_before$x[, 1], PC2 = pca_before$x[, 2],
batch = metadata$batch, condition = metadata$condition)
p1 <- ggplot(pca_df, aes(PC1, PC2, color = batch, shape = condition)) +
geom_point(size = 3) + ggtitle('Before Correction')
# PCA after correction
pca_after <- prcomp(t(corrected_expr), scale. = TRUE)
pca_df_after <- data.frame(PC1 = pca_after$x[, 1], PC2 = pca_after$x[, 2],
batch = metadata$batch, condition = metadata$condition)
p2 <- ggplot(pca_df_after, aes(PC1, PC2, color = batch, shape = condition)) +
geom_point(size = 3) + ggtitle('After Correction')
library(patchwork)
p1 + p2
Quantify Batch Effect
Goal: Measure the proportion of variance attributable to batch versus biological condition.
Approach: Correlate principal components with batch and condition labels, or use PVCA.
# PVCA - Principal Variance Component Analysis
library(pvca)
# Proportion of variance explained by batch vs condition
pvcaObj <- pvcaBatchAssess(normalized_expr, metadata, threshold = 0.6,
theInteractionTerms = c('batch', 'condition'))
# Or manual approach
pca <- prcomp(t(normalized_expr), scale. = TRUE)
variance_explained <- summary(pca)$importance[2, 1:5]
# Correlation of PCs with batch
cor(pca$x[, 1], as.numeric(as.factor(metadata$batch)))
Harmony (Single-Cell Integration)
Goal: Integrate single-cell data from multiple batches into a shared embedding.
Approach: Apply Harmony to PCA embeddings to iteratively remove batch effects while preserving cell-type structure.
library(harmony)
library(Seurat)
# For single-cell data with multiple batches
seurat_obj <- RunHarmony(seurat_obj, group.by.vars = 'batch', reduction = 'pca',
dims.use = 1:30)
# Use harmony reduction for downstream
seurat_obj <- RunUMAP(seurat_obj, reduction = 'harmony', dims = 1:30)
seurat_obj <- FindNeighbors(seurat_obj, reduction = 'harmony', dims = 1:30)
When NOT to Correct
# DON'T use batch-corrected values for:
# - Differential expression (use design formula instead)
# - Count-based methods expecting raw/normalized counts
# DO use batch-corrected values for:
# - Visualization (PCA, UMAP, heatmaps)
# - Clustering
# - Machine learning features
# - Cross-study comparisons
Related Skills
- differential-expression/deseq2-basics - DE with batch in design
- single-cell/clustering - Integration methods
- expression-matrix/matrix-operations - Data transformation
Related Skills
taste-skill
116.8kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.1kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.3kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.6k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
