The scoring system of our approach is a simple heuristic which can be improved over time as more functional SNVs are validated. Our analysis method and database are centered around a likely disruption of a protein–DNA interaction and/or expression. Therefore, a majority of our confidence in scoring comes from TF ChIP-seq data, particularly when a TF matches a PWM in the same context. We also derive confidence from SNVs overlapping eQTLs which have been shown to alter expression. It is likely that there are additional sources of data that reinforce each other in a different manner which should be explored. However, as shown when comparing to GWAS-lead and linkage SNVs, the scoring system provides significant enrichment concurrent with better category scores. As such, we believe that the database and scoring scheme provide the best current system for annotating and prioritizing variants.
