Format

Send to

Choose Destination
Nat Commun. 2018 Jan 30;9(1):441. doi: 10.1038/s41467-017-02806-4.

Probability of phenotypically detectable protein damage by ENU-induced mutations in the Mutagenetix database.

Author information

1
Center for the Genetics of Host Defense, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA. tao.wang@utsouthwestern.edu.
2
Quantitative Biomedical Research Center, Department of Clinical Science, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA. tao.wang@utsouthwestern.edu.
3
Kidney Cancer Program, Simmons Comprehensive Cancer Center, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA. tao.wang@utsouthwestern.edu.
4
Center for the Genetics of Host Defense, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA.
5
Quantitative Biomedical Research Center, Department of Clinical Science, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA.
6
Department of Statistical Science, Southern Methodist University, Dallas, TX, 75205, USA.
7
Immunology Division, Garvan Institute for Medical Research, Sydney, NSW, 2010, Australia.
8
Center for the Genetics of Host Defense, University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA. bruce.beutler@utsouthwestern.edu.

Abstract

Computational inference of mutation effects is necessary for genetic studies in which many mutations must be considered as etiologic candidates. Programs such as PolyPhen-2 predict the relative severity of damage caused by missense mutations, but not the actual probability that a mutation will reduce/eliminate protein function. Based on genotype and phenotype data for 116,330 ENU-induced mutations in the Mutagenetix database, we calculate that putative null mutations, and PolyPhen-2-classified "probably damaging", "possibly damaging", or "probably benign" mutations have, respectively, 61%, 17%, 9.8%, and 4.5% probabilities of causing phenotypically detectable damage in the homozygous state. We use these probabilities in the estimation of genome saturation and the probability that individual proteins have been adequately tested for function in specific genetic screens. We estimate the proportion of essential autosomal genes in Mus musculus (C57BL/6J) and show that viable mutations in essential genes are more likely to induce phenotype than mutations in non-essential genes.

PMID:
29382827
PMCID:
PMC5789985
DOI:
10.1038/s41467-017-02806-4
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Nature Publishing Group Icon for PubMed Central
Loading ...
Support Center