This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Variant Effect Prediction for Custom Genomes (A.Fumigatus)

Dear all,

currently, I am analyzing variants in genomes of Aspergillus Fumigatus strains and would like to predict the consequences of amino acid changes/frameshifts etc. For human genome data I am familiar with SIFT/PolyPhen etc. and routinely use that for annotation using ANNOVAR.

So far, I couldn't find any preconfigured databases for A.Fumigatus and tried to build my own genome-scale variant predictor database with PROVEAN (though I would be happy with any prediction tool). As I understand, I would have to generate a list with every single possible nucleotide exchange in the genome, but it is not very clear to me how to get to that point. Ideally, the output would be a list that could be used by ANNOVAR for annotation. Maybe someone else already ran into these issues with custom genomes and could share some knowledge?

Thanks!

genome sequencing
• 2,985 views • link
updated by Ram Ram • written by schmoo • schmoo

Thanks for the feedback Charles!

I already worked with SnpEff and it does annotate variants nicely. But my problem is rather to create a protein effect database that can be used by SnpEff/ANNOVAR to annotate a given (amino acid changing) variant with the predicted effect on the protein, e.g. damaging/deletorious/tolerable. For human genomes this is not a problem since PolyPhen/SIFT databases are readily available, but this is not the case for A.Fumigatus (as far as i know), meaning I would have to create such database from scratch. I am currently stuck at this step :(

• 0 views • link
written by schmoo • schmoo

Thanks for the hint! I already checked VEP from Ensembl but it seems to lack any information about PolyPhen/SIFT etc. (or I couldn't find it?). However, I found SIFT databases for A.fumigatus on (surprise) the SIFT homepage which can be easily adapted for annotation using ANNOVAR:

http://sift-db.bii.a-star.edu.sg/public/Aspergillus_fumigatus/

• 0 views • link
written by schmoo • schmoo

1 answer

I think SnpEff can work with any genome. It has fewer annotations than ANNOVAR, but I think it may be a better option in your case.

• 0 views • link
updated by Ram Ram • written by Charles Warden Charles Warden

Log in to answer this question.

More posts like this