fuNTRp: Identifying protein positions for variation driven functional tuning

Abstract

Motivation Evaluating the impact of non-synonymous genetic variants is essential for uncovering disease associations. Understanding the corresponding changes in protein sequences can also help with synthetic protein design and stability assessments. Even though hundreds of computational approaches addressing this task exist, and more are being developed, there has been little improvement in their performance in the recent years. One of the likely reasons for this lack of progress might be that most approaches use similar sets of gene/protein features for model development, with great emphasis being placed on sequence conservation. While high levels of conservation clearly highlight residues essential for protein activity, much of the in vivo observable variation is arguably weaker in its impact and, thus, requires evaluation of a higher level of resolution.Results Here we describe function Neutral/Toggle/Rheostat predictor (funtrp), a novel computational method that classifies protein positions by type based on the expected range of mutational impacts at that position: Neutral (most mutations have no or weak effects), Rheostat (range of effects; i.e. functional tuning), or Toggle (mostly strong effects). Three conclusions of our work are most salient. We show that our position types do not correlate strongly with the familiar protein features such as conservation or protein disorder. Moreover, we find that position type distribution varies across different enzyme classes. Finally, we demonstrate that position types reflect experimentally derived functional effects, improving performance of existing variant effect predictors and suggesting a way forward for the development of new ones.Availability https://services.bromberglab.org/funtrp; Git: https://bitbucket.org/bromberglab/funtrp/Contact mmilleratbromberglab.orgSupplementary information Supplementary data are available online.

Publication
bioRxiv
Avatar
Maximilian Miller
PostDoctoral Associate

improving variant effect predictions, deciphering metagenomes and creating an efficient service infrastructure