Almost all locations are called by samtools from RNA-seq data

[Younghoon Kim]

This is the first writing here. (So far I have only observed many threads here.) Please let me have the solution for this issue.


From RNA-seq, I have tried to call SNVs using samtools. (mpileup)
Previously, when I used Mapsplice as an alighment tool, the SNV calling using Samtools worked well. However, now I am trying to use RNASEQR alignment tool (NAR 2012; this is shown to be good for SNV calling purpose; this tool mainly use bowtie-alignment inside), and the Samtools results (using same parameter as for mapsplice) from the RNASEQR-generated sam/bam files were strange. The vcf file size is big, and when I look at inside the files, almost all locations are called as SNVs.


as a short example,

1 14544 . T C 6.02 ...
1 14545 . T C 6.02 ...
1 14546 . C A 6.02 ...
1 14547 . C G 6.02 ...
1 14548 . G C 6.02 ...
1 14551 . C T 6.02 ...


Do you know what is the reason of this? Please let me solve it.


Thanks,

[swbarnes2]

It looks like there is a mismatch between your .bam and your reference file. I'd double-check that you are using the exact same reference file for the alignment as you are using in mpileup. Then I'd remake the reference index file like this

samtools faidx reference.fa

Check to make sure that the command completed with no errors. Then rerun mpileup.

[Younghoon Kim]

Thanks so much!! I applied your suggestion, and it worked in my cases. I really thank to you.