anntools

Reaching out to the author to get a response for questions about the program has been tough. The CNV calling tools will only report 1 gene per CNV, no matter what the size of the CNV. It is either a bug or I am missing something with the options. In either the case the documentation is sparse and there is no support provided by the author. I am going to use another tool.

AnnTools requires that one load a mysql database of gene annotations and provides scripts to annotate one's VCF files against it. The problem with this methodology is that if you have many deeply sequenced exomes/transcriptomes/genomes (as I do), you'll need to have dedicated, powerful machine(s) to run AnnTools since you'll be limited by the number of database connections you can make at any given time. Furthermore, I encountered a few cases where AnnTools reported that a particular variant was located in an intron and CDS, while other annotation databases, such as UCSC Genome Browser, reported the variant to only be located in an intron. Emails sent to the author requesting more information, as well as providing the potential source of the bug, were not answered which is especially frustrating since the online documentation is sparse. Since AnnTools was inefficient and produced suspicious results, I went looking for another annotation tool and, so far, I've been quite pleased with snpEff. Annotating my libraries with snpEff takes minutes as opposed to weeks/months with AnnTools.