Extracting Reads from a BAM/SAM file that map back to a subset of contigs

[ediane]

I had some environmental HiSeq data that I assembled into contigs and clustered into putative genomes. Now I wan't to go back into my BAM files and extract reads mapping back to the a list of contigs (e.g mapping back to contigs from one of my putative genomes) into a fastq file.

I can't find a way to do this though, anyone know of a way?

[GenoMax]

Option 1 (preferred): Use samtools view command as described in this thread: http://seqanswers.com/forums/showthread.php?t=71912

Option 2: You could also go back to the original data and use bbsplit.sh from BBMap to bin the reads?

[SNPsaurus]

bbduk will take the contigs as a reference and find reads that have kmers and create matching and non-matching files.

If you want to go from the bam, then samtools view extracts reads from a bam file, and you can specify chromosome (or contig) and position ranges. Then bbtools reformat to go from sam to fastq.