top of page
  • vitod1

MIS: A Multimodal Measure of Human Protein Intolerance to Functional Missense Variations

Taha Mohseni Ahooyi, PhD (1) Benjamin Stear, MSc (1) Deanne Taylor, PhD (1,2) (1) Department of Biological & Health Informatics (DBHi), the Children's Hospital of Philadelphia (2) Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania

Poster # 90

Single nucleotide missense variants are likely to introduce functional alterations to gene products with the severity of such events ranging from negligible effects (same level as synonymous variants) to the complete disruption of protein function. Some computational tools (such as SIFT) have been developed to assess genetic variation risk in silico, nevertheless the complex nature of protein functionality in the context of biological interactions has urged researchers to look beyond the simulations and incorporate information from the real-world studies. To this end, multiple measures of genic intolerance have been proposed based on large-scale human exome sequencing data such as ExAC and gnomAD, the majority of which utilize variant depletion metrics and the concept of negative selection to create estimates of protein and/or residue intolerance to functional variations. Although proven to be useful tools for interpretation, the current intolerance measures are impacted by inadequate sample sizes and oversimplifying assumptions, limiting their application to differentiate common variants form rare disease-causing variants with high certainty. To address some shortcomings associated with of these methods, here we propose a composite missense intolerance score (MIS) that incorporates information from orthogonal genetics data, i.e. exome sequencing-based measures (MTR v1 and v2 based on ExAC and gnomAD, respectively) as well as CADD scores and ClinVar genetic testing results, to generate missense pathogenicity scores at the protein and residue resolution levels. We provide quantitive comparison between MIS and some reference measures in differentiating known mendelian disease-causing genes (OMIM Morbid Map, ClinGen happloinsuffient and ClinVar genes with pathogenic variants) from their benign counterparts. Finally, we will provide a comprehensive list of high MIS proteins with little or no evidence of pathogenicity as potential candidates for further research and interpretations.

4 views0 comments

Recent Posts

See All


bottom of page