/bio-genome-intervals-bigwig-tracks

<!--

COPYRIGHT NOTICE

This file is part of the "Universal Biomedical Skills" project.

Copyright (c) 2026 MD BABU MIA, PhD <md.babu.mia@mssm.edu>

All Rights Reserved.

#

This code is proprietary and confidential.

Unauthorized copying of this file, via any medium is strictly prohibited.

#

Provenance: Authenticated by MD BABU MIA

-->

--- name: bio-genome-intervals-bigwig-tracks description: Create and read bigWig browser tracks for visualizing continuous genomic data. Convert bedGraph to bigWig, extract signal values, and generate coverage tracks using UCSC tools and pyBigWig. Use when preparing coverage tracks for genome browsers or extracting signal at specific regions. tool_type: mixed primary_tool: pyBigWig measurable_outcome: Execute skill workflow successfully with valid output within 15 minutes. allowed-tools:

• read_file
• run_shell_command

---

BigWig Tracks

BigWig is an indexed binary format for continuous genomic data. Efficient for genome browsers and programmatic access.

Why BigWig?

| Format | Size | Random Access | Browser Support | |--------|------|---------------|-----------------| | bedGraph | Large | No | Limited | | bigWig | ~10x smaller | Yes (indexed) | Excellent |

Convert bedGraph to bigWig (CLI)

Installation

# UCSC tools
conda install -c bioconda ucsc-bedgraphtobigwig ucsc-bigwigtobedgraph

# Or download directly
wget http://hgdownload.soe.ucsc.edu/admin/exe/linux.x86_64/bedGraphToBigWig
chmod +x bedGraphToBigWig

Basic Conversion

# Sort bedGraph first (required)
sort -k1,1 -k2,2n coverage.bedGraph > coverage.sorted.bedGraph

# Convert to bigWig
bedGraphToBigWig coverage.sorted.bedGraph chrom.sizes output.bw

# chrom.sizes format: chr<TAB>size
# chr1	248956422
# chr2	242193529

Get Chromosome Sizes

# From FASTA index
cut -f1,2 reference.fa.fai > chrom.sizes

# Download from UCSC
wget https://hgdownload.soe.ucsc.edu/goldenPath/hg38/bigZips/hg38.chrom.sizes

# From BAM header
samtools view -H alignments.bam | grep @SQ | sed 's/@SQ\tSN:\|LN://g' > chrom.sizes

Full Workflow

# Generate bedGraph from BAM
bedtools genomecov -ibam alignments.bam -bg > coverage.bedGraph

# Sort bedGraph
sort -k1,1 -k2,2n coverage.bedGraph > coverage.sorted.bedGraph

# Convert to bigWig
bedGraphToBigWig coverage.sorted.bedGraph hg38.chrom.sizes coverage.bw

# Clean up intermediate files
rm coverage.bedGraph coverage.sorted.bedGraph

Read BigWig with pyBigWig (Python)

Installation

pip install pyBigWig

Open and Inspect

import pyBigWig

# Open file
bw = pyBigWig.open('coverage.bw')

# File info
print(f'Chromosomes: {bw.chroms()}')
print(f'Header: {bw.header()}')

# Check if file is bigWig (not bigBed)
print(f'Is bigWig: {bw.isBigWig()}')

# Close when done
bw.close()

Extract Values

import pyBigWig

bw = pyBigWig.open('coverage.bw')

# Get values for a region (returns numpy array)
values = bw.values('chr1', 1000000, 1001000)
print(f'Mean: {values.mean():.2f}')
print(f'Max: {values.max():.2f}')

# Get specific intervals with values
intervals = bw.intervals('chr1', 1000000, 1001000)
# Returns: [(start, end, value), ...]
for start, end, val in intervals:
    print(f'{start}-{end}: {val}')

# Statistics for region
stats = bw.stats('chr1', 1000000, 1001000, type='mean')
print(f'Mean coverage: {stats[0]:.2f}')

# Available stat types: mean, min, max, coverage, std, sum
max_val = bw.stats('chr1', 1000000, 1001000, type='max')
coverage = bw.stats('chr1', 1000000, 1001000, type='coverage')

bw.close()

Binned Statistics

import pyBigWig

bw = pyBigWig.open('coverage.bw')

# Get mean values in 100bp bins across region
region_start, region_end = 1000000, 2000000
n_bins = 1000  # 100bp bins

binned = bw.stats('chr1', region_start, region_end, type='mean', nBins=n_bins)
# Returns list of n_bins values

bw.close()

Extract for BED Regions

import pyBigWig
import pybedtools

bw = pyBigWig.open('coverage.bw')
bed = pybedtools.BedTool('regions.bed')

# Get mean signal per region
results = []
for interval in bed:
    chrom, start, end = interval.chrom, interval.start, interval.end
    mean_signal = bw.stats(chrom, start, end, type='mean')[0]
    results.append({
        'chrom': chrom,
        'start': start,
        'end': end,
        'name': interval.name,
        'signal': mean_signal if mean_signal else 0
    })

bw.close()

# Convert to DataFrame
import pandas as pd
df = pd.DataFrame(results)
print(df)

Create BigWig with pyBigWig

import pyBigWig

# Create new bigWig
bw = pyBigWig.open('output.bw', 'w')

# Add header (chromosome sizes)
bw.addHeader([('chr1', 248956422), ('chr2', 242193529)])

# Add entries (must be sorted by position)
# Method 1: Individual entries
bw.addEntries(['chr1', 'chr1'], [0, 100], ends=[100, 200], values=[1.5, 2.3])

# Method 2: Chromosome at a time (more efficient)
bw.addEntries('chr1', [0, 100, 200], ends=[100, 200, 300], values=[1.5, 2.3, 3.1])

# Method 3: Fixed-width spans (most efficient for dense data)
bw.addEntries('chr2', 0, values=[1.0, 2.0, 3.0, 4.0], span=100, step=100)
# Creates: chr2:0-100=1.0, chr2:100-200=2.0, chr2:200-300=3.0, chr2:300-400=4.0

bw.close()

deepTools for BigWig Operations

Installation

conda install -c bioconda deeptools

Generate Normalized BigWig from BAM

# RPKM normalization
bamCoverage -b alignments.bam -o coverage.bw --normalizeUsing RPKM

# CPM normalization
bamCoverage -b alignments.bam -o coverage.bw --normalizeUsing CPM

# BPM (bins per million) - like TPM for ChIP-seq
bamCoverage -b alignments.bam -o coverage.bw --normalizeUsing BPM

# With bin size and smoothing
bamCoverage -b alignments.bam -o coverage.bw \
    --binSize 10 \
    --normalizeUsing CPM \
    --smoothLength 30

# Extend reads to fragment length
bamCoverage -b alignments.bam -o coverage.bw \
    --extendReads 200 \
    --normalizeUsi