Need to account for IonTorrent sequencing artifacts; multiple variants at locus

[stevekm]

Sometimes a variant is called in a sample, but similar variants are also present in the control sample for the run. Consider one or more of the following:

• for every sample, look for similar variants in control sample
• for IonTorrent VCF variant entries that have multiple called variants at a single location, develop system of flagging these variants as possible artifacts
• The default ~500 height IGV snapshots might not be tall enough to see the presence of indels and other artifacts throughout the reads at a locus, consider generating extra long ~5000 height snapshots for these variants; link to report?
• need to develop system for quickly & easily removing these from the report & summary tables upon manual review and feedback from clinicians

[stevekm]

Issue resolved with the addition of long snapshots in IGV for low frequency variants