I uploaded a standard VCF that contains 57651 lines (or 57471 lines variant data) and ran it with ANNOVAR , under usegalaxy.org, to annotate the variants. To my surprise, the output from ANNOVAR contains only 24072 variants. When I aligned the two tables, it's clear ANNOVAR "removed" some of the variants (both novel and known SNPs, granted ANNOVAR on Galaxy only uses snp137) in the original VCF.
Is there some default setting in ANNOVAR that makes some cutoff for variant annotation ? I don't see how I can change any parameter when running ANNOVAR on usegalaxy. Any explanation for the short list of ourput or any suggestion what I could do to re-annotate the VCF? Great many thanks.