Indexing with bwa and samtools

[scami]

Hi guys!

I indexed a reference genome with the bwa index command. I then aligned a fastq file on it and obtained several gigabytes sam alignment file. When I use the "samtools mpileup" command I get only few thousands of bases aligned. For the samtools command I use the reference genome that I indexed with bwa (by the command bwa index referenceGenome.fasta). Is this ok? Or do I need to index the reference genome again by using the command samtools faidx?
Thanks for your help!

[Brian Bushnell]

To map with bwa, you index with bwa, not samtools, so what you did should be right. However, it's possible you used the wrong reference for those reads. The size of the sam file is related to the amount of input, not the amount that mapped, so an incredibly low rate of mapping normally indicates you're mapping to the wrong reference.

I suggest you blast a few of the reads to see what they came from.

And by the way, it's always a good idea to post your exact command lines and the exact output when you encounter a problem.

[mastal]

Run samtools flagstat to give you some basic stats on your alignment, like what percentage of the reads aligned.

You should use the same reference genome that you used to make the bwa index, but not the bwa index, just the reference fasta file.

And your bam alignment file needs to be sorted and indexed with samtools as well.

[scami]

Thanks for the answer. Your suggestion led me to the right way. As a matter of fact I used fastq-dump to extract the sra sequence file. Although they were paired end I did not use the --split-files tag since I thought that would just divide the ends in two different files. However extracting paired end sra files without that flag creates fastq files in which both end are reported as a unique read........
this was evident when blasting the reads because only half aligned to the reference genome
thanks!