Entering edit mode
16 months ago
Ed
•
0
I tried to build a local database of bovine ARS-UCD1.2 using SNPeff, and in my annotated results, the SNPS of introns were always the most numerous, and I was very confused
Introns comprise a larger percent of the genome than exons, so intronic SNVs being more numerous than exonic SNVs makes sense.
Thank you! But shouldn't intergenic regions have the most SNVs?
Fair point. It also depends on the annotations, I think. Do you see other non-coding annotations such as upstream, downstream, etc.?
yes, I get. I think this is a problem of snpeff, and maybe the annotation of the bovine genome also influenced the results. For the past year, I have had to use Annovar software to get accurate annotation results. I don't know how to solve it. Can you offer any advice? Thank you very much
I find VEP much better than both snpEff and Annovar, but I work on the human genome. Is there a reason you're looking to switch from Annovar given it provides you with accurate results?
Thank you, I will try VEP, I think snpEff is more convenient to use.
It definitely is quite convenient especially compared to VEP's complicated options but VEP is far more detailed plus EnsEMBL is much better annotated than RefSeq IMO but again, I speak from a human context. This may not apply to the bovine genome.