Variant call using BAM from Tophat

[emilyjia2000]

Hello,

I used tophat on mapping of RNA-seq samples, after running TopHat I got accepted_hit.bam file. Is it possible to use this BAM on the variant call analysis using Samtools? I overheard that BAM from TopHat is not good to variant call, but I don't know the reason. Which are alignment tools good for RNA-seq to do the variant call?

many thanks,

[chadn737]

I would also be interested in this answer as well.

I wonder if it has anything to do with the fact that Tophat does not take into account quality scores while doing alignments, whereas aligners like MAQ do? However MAQ I think can't handle splicing, so is not a very good option for RNA-seq.

[emilyjia2000]

By the way, the data was process by Illumina, does Eland alignment work well on variant call? or BWA?

[emilyjia2000]

chadn737,
If that is the case, is it possible to use samtools filter to filter out the lower quality score?

[chadn737]

I have no experience with ELAND. However, you should be able to filter out low quality alignments with samtools, so one approach would be to use Tophat, make variant calls, and then filter.

[honey]

TopHAT

I believe TopHat is manily align at junctions with the help of Bowtie? How about if you use Botiwe an duse samtools. I am not very sure?