A deep learning framework for predicting human essential genes from population and functional genomic data

Being able to predict essential genes intolerant to loss-of-function (LOF) mutations can dramatically improve our ability to identify genes associated with genetic disorders. Numerous computational methods have recently been developed to predict human essential genes from population genomic data; ho...

Full description

Saved in:
Bibliographic Details
Published inbioRxiv
Main Authors Lapolice, Troy M, Yi-Fei, Huang
Format Paper
LanguageEnglish
Published Cold Spring Harbor Cold Spring Harbor Laboratory Press 23.12.2021
Subjects
Cell lines
Computer applications
Deep learning
Genetic disorders
Genomics
Online AccessGet full text

Cover

More Information
Summary:Being able to predict essential genes intolerant to loss-of-function (LOF) mutations can dramatically improve our ability to identify genes associated with genetic disorders. Numerous computational methods have recently been developed to predict human essential genes from population genomic data; however, the existing methods have limited power in pinpointing short essential genes due to the sparsity of polymorphisms in the human genome. Here we present an evolution-based deep learning model, DeepLOF, which integrates population and functional genomic data to improve gene essentiality prediction. Compared to previous methods, DeepLOF shows unmatched performance in predicting ClinGen haploinsufficient genes, mouse essential genes, and essential genes in human cell lines. Furthermore, DeepLOF discovers 109 potentially essential genes that are too short to be identified by previous methods. Altogether, DeepLOF is a powerful computational method to aid in the discovery of essential genes. Competing Interest Statement The authors have declared no competing interest. Footnotes * https://github.com/yifei-lab/DeepLOF
DOI:10.1101/2021.12.21.473690