countmut 0.0.3

Ultra-fast strand-aware mutation counter

CountMut

Ultra-fast strand-aware mutation counter for bisulfite sequencing analysis

CountMut is a high-performance tool for counting mutations from bisulfite sequencing BAM files (BS-seq, CAM-seq, GLORI-seq, eTAM-seq). It features parallel processing, quality-based mate overlap deduplication, and optimized file I/O for maximum speed.

Features

• 🚀 Ultra-Fast: Call mutation without pileup reads
• 🧬 Bisulfite Support: NS, Zf, Yf tag filtering for conversion analysis
• 🎯 Accurate: Quality-based mate overlap deduplication prevents double-counting
• ⚡ Parallel: Multi-threaded genomic window processing
• 🔧 Flexible: Configurable filtering, strand-specific processing, auto-indexing

Installation

pip install countmut

Quick Start

# Basic usage - auto-creates indices if needed
countmut -i input.bam -r reference.fa -o mutations.tsv

# Count T→C mutations (common in bisulfite sequencing)
countmut -i input.bam -r reference.fa -o mutations.tsv --ref-base T --mut-base C

# With custom threads and filtering
countmut -i input.bam -r reference.fa -o mutations.tsv -t 8 --max-unc 5 --min-con 2

Options

Input/Output

-i, --input PATH       Input BAM file (coordinate-sorted) [required]
-r, --reference PATH   Reference FASTA file [required]
-o, --output PATH      Output TSV file (default: stdout)
-f, --force            Overwrite output without prompting

Mutation Analysis

--ref-base TEXT        Reference base to count from [default: A]
--mut-base TEXT        Mutation base to count [default: G]
--strand TEXT          Strand: both/forward/reverse [default: both]
--region TEXT          Genomic region (e.g., 'chr1:1000000-2000000')

Performance

-t, --threads INTEGER  Number of parallel threads [default: auto]
-b, --bin-size INTEGER Genomic bin size in bp [default: 10000]

Quality Filters

--min-baseq INTEGER    Min base quality (Phred score) [default: 20]
--min-mapq INTEGER     Min mapping quality (MAPQ) [default: 0]
--max-sub INTEGER      Max substitutions (NS tag) [default: 1]
--trim-start INTEGER   Trim 5' bases from reads [default: 2]
--trim-end INTEGER     Trim 3' bases from reads [default: 2]
--max-unc INTEGER      Max unconverted (Zf tag) [default: 3]
--min-con INTEGER      Min converted (Yf tag) [default: 1]

Output Records

-p, --pad INTEGER      Motif window half-size [default: 15]
-s, --save-rest        Include other bases (o0, o1, o2 columns)

Note: BAM files must have NS, Zf, and Yf tags (essential for bisulfite analysis). Indices (.bai, .fai) are created automatically if missing.

Output Format

TSV file with the following columns:

Column	Description
chrom	Chromosome name
pos	Genomic position (1-based)
strand	Strand (+ or -)
motif	Sequence context (2×pad+1 bp window)
u0, u1, u2	Unconverted (reference base) counts
m0, m1, m2	Mutation (mutation base only) counts
o0, o1, o2	Other bases counts (with --save-rest)

Count categories (x0, x1, x2):

• x0 (low quality): Bases failing quality filters (trim region, max-sub, min-mapq, min-baseq)
• x1 (high conversion): Bases from reads with high conversion efficiency (low Zf and high Yf)
• x2 (insufficient conversion): Bases from reads with insufficient conversion efficiency (high Zf or low Yf)

 

Copyright © 2025-present Chang Y