BWA-MEM2

Overview

BWA-MEM2 is a highly optimised re-implementation of the BWA-MEM short-read aligner. It uses SIMD (Single Instruction, Multiple Data) acceleration to deliver significantly faster alignment speeds while producing identical output to the original BWA-MEM. BWA-MEM2 is the recommended aligner for mapping Illumina whole-genome sequencing, whole-exome sequencing, and other short-read data to a reference genome.

Installation

mamba install -c bioconda bwa-mem2

Basic Usage

Index the reference genome, align paired-end reads, and sort the output into a coordinate-sorted BAM file.

# Index reference genome
bwa-mem2 index /ref/GRCh38.fa

# Align paired-end reads
bwa-mem2 mem -t 16 /ref/GRCh38.fa \
  sample_R1.fastq.gz sample_R2.fastq.gz \
  | samtools sort -@ 4 -o sample.sorted.bam -

samtools index sample.sorted.bam

Key Parameters

Flag / option	Description
`-t`	Number of alignment threads (default 1).
`-R`	Read group header line (e.g., `'@RG\tID:sample\tSM:sample\tPL:ILLUMINA'`). Required for downstream variant calling.
`-M`	Mark shorter split hits as secondary (for Picard compatibility).
`-k`	Minimum seed length (default 19). Lower values increase sensitivity at the cost of speed.
`-w`	Band width for banded alignment (default 100).
`-A`	Matching score (default 1).
`-B`	Mismatch penalty (default 4).
`-O`	Gap open penalty (default 6,6).
`-E`	Gap extension penalty (default 1,1).
`-Y`	Use soft clipping for supplementary alignments.

Expected Output

The pipeline above produces:

sample.sorted.bam – a coordinate-sorted BAM file containing all aligned and unaligned reads.
sample.sorted.bam.bai – the BAM index file created by samtools index.

Verify the alignment with samtools flagstat:

samtools flagstat sample.sorted.bam

This prints the total number of reads, mapped reads, properly paired reads, and other alignment statistics.