This is an old revision of the document!
====== Overview ====== BWA is a tool for aligning short reads to a reference genome using two different algorithms based on the Burrows-Wheeler Transform (BWT). ====== Install ====== The source for BWA can be found on the [[http://bio-bwa.sourceforge.net/| BWA SourceForge page]]. The archive (v0.5.9) was extracted to the ''/campusdata/BME235/programs'' directory, compiled and the executable was copied to ''/campusdata/BME235/bin''. ====== Using BWA ====== The first step in using BWA is building a reference database. <code> bwa index -a is database.fasta </code> Their are two options for the algorithm. The default option, ''is'', is relatively fast and works on genomes smaller than 2GB. The other algorithm, ''bwtsw'', is slower and less accurate but works on longer reads and works with larger databases. Next, the reads are aligned to the reference using the ''aln'' command. <code> bwa aln database.fasta short_read.fastq > aln_sa.sai </code> The reads in the fastq file are aligned against the reference data base and the results are written to standard output in the ''.sai'' format. The 'samse' and 'sampe' commands are used to generate single-end and paired-end alignments in SAM format from the FASTQ and SAI files. <code> bwa samse database.fasta aln_sa.sai short_read.fastq > aln.sam bwa sampe database.fasta aln_sa1.sai aln_sa2.sai read1.fq read2.fq > aln.sam </code>
