Computational assessment of feature combinations for pathogenic variant prediction

ConclusionExisting pathogenicity prediction methods can be further improved.

We investigated which combination of existing methods, variant features, and gene features archived the best performance in the prediction of pathogenic variants. We found that combining the number of “biological process” Gene Ontology annotations of a gene with the methods PON‐P2 and PROVEAN significantly improved prediction of pathogenic variants, outperforming all individual methods. Analysis of the Gene Ontology feature suggests that it is not a variant‐dependent annotation bias but reflects the multifunctional nature of disease genes.

Source: Molecular Genetics & Genomic Medicine - March 14, 2016 Category: Genetics & Stem Cells Authors: Eva König, Johannes Rainer, Francisco S. Domingues Tags: Original Article Source Type: research