Calling SNP in RNA seq

[figo1019]

Hi All

I have assembled two mammalian transcriptome which doesn't have reference genome. I have assembled it using Trinity and now I have used varscan to call the SNP by mapping the reads to the contigs.
I have question that can I merge all the reads of both the individual
and make a trinity assembly out of that and later map back the individual reads to that to call the SNP in each individual. Is it right way to combine the reads and call SNP of same species individual. These two transcriptome is from same species and same tissues ( 2 individuals).

Best

[Brian Bushnell]

There's no single "right way", but there are a lot of wrong ways. Your proposal sounds fine. Assemblies from multiple individuals may be slightly worse than assemblies from a single individual due to polymorphism, but mammals are diploid anyway, so presumably the assembler can deal with some polymorphism. You'll probably end up with a more complete transcriptome by assembling both together.

Of course, the mapping and variant-calling should be done individually.