bio-chipseq-visualization

## Version Compatibility Reference examples tested with: GenomicRanges 1.54+, deepTools 3.5+ Before using code patterns, verify installed versions match. If versions differ: - R: `packageVersion('<pkg>')` then `?function_name` to verify parameters - CLI: `<tool> --version` then `<tool> --help` to confirm flags If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying. # ChIP-seq Visualization **"Create a heatmap of ChIP-seq signal around peaks"** → Generate signal heatmaps, profile plots, and genome browser tracks showing enrichment patterns around genomic features. - CLI: `deeptools computeMatrix reference-point` → `plotHeatmap` - R: `Gviz`, `ChIPseeker::plotAvgProf()` ## deepTools - Compute Matrix **Goal:** Build a signal matrix of ChIP-seq coverage around reference points for downstream heatmaps and profiles. **Approach:** Use computeMatrix to extract bigWig signal values in windows around genomic features like TSS. ```bash # Compute signal matrix around TSS computeMatrix reference-point \ --referencePoint TSS \ -b 3000 -a 3000 \ # 3kb upstream and downstream -R genes.bed \ # Reference regions -S sample.bw \ # Signal file (bigWig) -o matrix.gz \ --outFileSortedRegions sorted_genes.bed ``` ## deepTools - Scale-Regions **Goal:** Visualize ChIP signal across gene bodies scaled to a uniform length. **Approach:** Scale all gene regions to equal size and compute signal with flanking windows. ```bash # Signal across gene bodies computeMatrix scale-regions \ -R genes.bed \ -S sample1.bw sample2.bw \ -b 3000 -a 3000 \ # Flanking regions -m 5000 \ # Scaled body length -o matrix_scaled.gz ``` ## deepTools - Heatmap **Goal:** Generate a heatmap of ChIP-seq signal intensity across genomic regions. **Approach:** Render the precomputed signal matrix as a clustered heatmap with optional profile summary. ```bash # Generate heatmap from matrix plotHeatmap \ -m matrix.gz \ -o heatmap.png \ --colorMap RdBu \ --whatToShow 'heatmap and colorbar' \ --zMin -3 --zMax 3 # With profile on top plotHeatmap \ -m matrix.gz \ -o heatmap_with_profile.png \ --plotTitle 'H3K4me3 Signal' \ --heatmapHeight 15 \ --refPointLabel TSS ``` ## deepTools - Profile Plot **Goal:** Display average ChIP-seq signal profiles across genomic regions for sample comparison. **Approach:** Plot mean signal from the computed matrix, optionally overlaying multiple samples. ```bash # Average profile plot plotProfile \ -m matrix.gz \ -o profile.png \ --plotTitle 'Average Signal Profile' \ --perGroup # Multiple samples comparison plotProfile \ -m matrix_multi.gz \ -o profile_compare.png \ --colors red blue green \ --plotTitle 'Sample Comparison' ``` ## Create BigWig from BAM **Goal:** Convert BAM alignments to normalized bigWig signal tracks for visualization. **Approach:** Use bamCoverage for single-sample normalization or bamCompare for log2 ratio of ChIP over input. ```bash # Normalized bigWig (CPM) bamCoverage \ -b sample.bam \ -o sample.bw \ --normalizeUsing CPM \ --binSize 10 \ --numberOfProcessors 8 # With input subtraction bamCompare \ -b1 chip.bam \ -b2 input.bam \ -o chip_vs_input.bw \ --operation log2ratio \ --binSize 50 ``` ## ChIPseeker Profile Heatmap (R) **Goal:** Visualize peak distribution around TSS using ChIPseeker tag matrices and profile plots. **Approach:** Build a tag density matrix from peak locations relative to promoter windows, then plot as heatmap or average profile. ```r library(ChIPseeker) library(TxDb.Hsapiens.UCSC.hg38.knownGene) txdb <- TxDb.Hsapiens.UCSC.hg38.knownGene # Load peaks peaks <- readPeakFile('sample_peaks.narrowPeak') # Get promoter regions promoter <- getPromoters(TxDb = txdb, upstream = 3000, downstream = 3000) # Compute tag matrix tagMatrix <- getTagMatrix(peaks, windows = promoter) # Heatmap tagHeatmap(tagMatrix, xlim = c(-3000, 3000), color = 'red') # Profile plot plotAvgProf(tagMatrix, xlim = c(-3000, 3000), xlab = 'Distance from TSS (bp)', ylab = 'Peak Count Frequency') # With confidence interval plotAvgProf2(tagMatrix, xlim = c(-3000, 3000), conf = 0.95) ``` ## Gviz - Genome Browser Tracks (R) **Goal:** Create publication-quality genome browser views combining signal tracks, gene models, and ideograms. **Approach:** Layer Gviz track objects (ideogram, axis, data, gene) and render a specific genomic region. ```r library(Gviz) library(GenomicRanges) # Define region chr <- 'chr1' start <- 1000000 end <- 1100000 # Ideogram track itrack <- IdeogramTrack(genome = 'hg38', chromosome = chr) # Genome axis gtrack <- GenomeAxisTrack() # Data track from bigWig dtrack <- DataTrack( range = 'sample.bw', genome = 'hg38', type = 'histogram', name = 'ChIP Signal', col.histogram = 'darkblue', fill.histogram = 'darkblue' ) # Gene track library(TxDb.Hsapiens.UCSC.hg38.knownGene) txdb <- TxDb.Hsapiens.UCSC.hg38.knownGene grtrack <- GeneRegionTrack(txdb, genome = 'hg38', chromosome = chr, name = 'Genes') # Plot plotTracks(list(itrack, gtrack, dtrack, grtrack), from = start, to = end, chromosome = chr) ``` ## Multiple Samples in Gviz **Goal:** Compare ChIP-seq signal from multiple samples in a single browser view. **Approach:** Create separate DataTrack objects per sample and stack them in the plotTracks call. ```r # Create data tracks for each sample dtrack1 <- DataTrack(range = 'control.bw', genome = 'hg38', name = 'Control', type = 'histogram', col.histogram = 'blue', fill.histogram = 'blue') dtrack2 <- DataTrack(range = 'treatment.bw', genome = 'hg38', name = 'Treatment', type = 'histogram', col.histogram = 'red', fill.histogram = 'red') plotTracks(list(itrack, gtrack, dtrack1, dtrack2, grtrack), from = start, to = end, chromosome = chr) ``` ## EnrichedHeatmap (R) **Goal:** Generate customizable heatmaps of ChIP signal around genomic features using ComplexHeatmap framework. **Approach:** Normalize bigWig signal to a matrix around target sites and render with EnrichedHeatmap. ```r library(EnrichedHeatmap) library(rtracklayer) # Load signal and regions signal <- import('sample.bw') tss <- promoters(txdb, upstream = 0, downstream = 1) # Normalize to matrix mat <- normalizeToMatrix(signal, tss, extend = 3000, mean_mode = 'w0', w = 50) # Heatmap EnrichedHeatmap(mat, name = 'Signal', col = c('white', 'red')) ``` ## IGV Batch Screenshot **Goal:** Automate genome browser screenshots at specific loci without manual interaction. **Approach:** Write an IGV batch script that loads tracks, navigates to regions, and saves snapshots. ```bash # Create IGV batch script cat > igv_batch.txt << 'EOF' new genome hg38 load sample.bw load peaks.bed goto chr1:1000000-1100000 snapshot region1.png goto chr2:50000000-51000000 snapshot region2.png exit EOF # Run IGV in batch mode igv.sh -b igv_batch.txt ``` ## Key Tools Comparison | Tool | Type | Best For | |------|------|----------| | deepTools | CLI | Large-scale heatmaps, profiles | | ChIPseeker | R | Peak-centric visualization | | Gviz | R | Publication-quality browser | | EnrichedHeatmap | R | Customizable heatmaps | | IGV | GUI | Interactive exploration | ## deepTools Key Commands | Command | Purpose | |---------|---------| | bamCoverage | BAM to bigWig | | bamCompare | Compare two BAMs | | computeMatrix | Signal matrix | | plotHeatmap | Heatmap visualization | | plotProfile | Profile plot | | multiBigwigSummary | Compare multiple bigWigs | | plotCorrelation | Sample correlation | ## Related Skills - peak-calling - Generate peaks for visualization - peak-annotation - Annotation pie charts - alignment-files - Prepare BAM files