Hi all,
I have a BAM (paired end) file which I want to convert to individual chromosome fasta or fastq files. We tried to use EMBOSS 6.6 but this did not work. Can someone please help with how this can be accomplished.
thanks.
I'm not completely sure, if I got your question correctly, but using this code, you'll write out all reads that were mapped to chr1 in a fastq-file. To make it for the other chromosomes, you can just read in the header, loop over the chromosomes and change 'chr1' to a variable.
samtools view -ub INPUT.bam chr1 | samtools bam2fq - > chr1.fastq
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
just a follow up question. the bam is aligned to a certain reference file, if we generate individual chromosome bam files and the convert each bam to fastq and align the chromosome fastq to another reference i am going to mess things up, since mapping could be different i.e a sequence which mapped to chromosome 1 in one reference could map to another chromosome in another reference.
Well, that depends... are you talking about different species? Within one species the differences between the assemblies should not be that different. If you, for example, map your reads against hg18 and hg19, the best hits should show up on the same chromosomes. If you map them on hg19 and ce10, they will map on different chromosomes. ;)
Its going to be same species, human. Even if say some sequences remapped to another chromosome and then if i merged all the new BAMs the overall effect would be the same.
I would highly recommend you to map the reads against the complete genome and not chromosome by chromosome. A correct merging of all bams is a pain in the a...