{"title":"Interpreting cis-regulatory interactions from large-scale deep neural networks","authors":"Shushan Toneyan, Peter K. Koo","doi":"10.1038/s41588-024-01923-3","DOIUrl":null,"url":null,"abstract":"<p>The rise of large-scale, sequence-based deep neural networks (DNNs) for predicting gene expression has introduced challenges in their evaluation and interpretation. Current evaluations align DNN predictions with orthogonal experimental data, providing insights into generalization but offering limited insights into their decision-making process. Existing model explainability tools focus mainly on motif analysis, which becomes complex when interpreting longer sequences. Here we present <i>cis</i>-regulatory element model explanations (CREME), an in silico perturbation toolkit that interprets the rules of gene regulation learned by a genomic DNN. Applying CREME to Enformer, a state-of-the-art DNN, we identify <i>cis</i>-regulatory elements that enhance or silence gene expression and characterize their complex interactions. CREME can provide interpretations across multiple scales of genomic organization, from <i>cis</i>-regulatory elements to fine-mapped functional sequence elements within them, offering high-resolution insights into the regulatory architecture of the genome. CREME provides a powerful toolkit for translating the predictions of genomic DNNs into mechanistic insights of gene regulation.</p>","PeriodicalId":18985,"journal":{"name":"Nature genetics","volume":null,"pages":null},"PeriodicalIF":31.7000,"publicationDate":"2024-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Nature genetics","FirstCategoryId":"99","ListUrlMain":"https://doi.org/10.1038/s41588-024-01923-3","RegionNum":1,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GENETICS & HEREDITY","Score":null,"Total":0}
引用次数: 0
Abstract
The rise of large-scale, sequence-based deep neural networks (DNNs) for predicting gene expression has introduced challenges in their evaluation and interpretation. Current evaluations align DNN predictions with orthogonal experimental data, providing insights into generalization but offering limited insights into their decision-making process. Existing model explainability tools focus mainly on motif analysis, which becomes complex when interpreting longer sequences. Here we present cis-regulatory element model explanations (CREME), an in silico perturbation toolkit that interprets the rules of gene regulation learned by a genomic DNN. Applying CREME to Enformer, a state-of-the-art DNN, we identify cis-regulatory elements that enhance or silence gene expression and characterize their complex interactions. CREME can provide interpretations across multiple scales of genomic organization, from cis-regulatory elements to fine-mapped functional sequence elements within them, offering high-resolution insights into the regulatory architecture of the genome. CREME provides a powerful toolkit for translating the predictions of genomic DNNs into mechanistic insights of gene regulation.
期刊介绍:
Nature Genetics publishes the very highest quality research in genetics. It encompasses genetic and functional genomic studies on human and plant traits and on other model organisms. Current emphasis is on the genetic basis for common and complex diseases and on the functional mechanism, architecture and evolution of gene networks, studied by experimental perturbation.
Integrative genetic topics comprise, but are not limited to:
-Genes in the pathology of human disease
-Molecular analysis of simple and complex genetic traits
-Cancer genetics
-Agricultural genomics
-Developmental genetics
-Regulatory variation in gene expression
-Strategies and technologies for extracting function from genomic data
-Pharmacological genomics
-Genome evolution