Seitz, Evan E, McCandlish, David M, Kinney, Justin B, Koo, Peter K (November 2023) Interpreting cis -regulatory mechanisms from genomic deep neural networks using surrogate models. bioRxiv. (Submitted)
Preview |
PDF
2023_Seitz_Interpreting_cis_regulatory_mechanisms_preprint.pdf - Submitted Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (7MB) | Preview |
Abstract
Deep neural networks (DNNs) have greatly advanced the ability to predict genome function from sequence. Interpreting genomic DNNs in terms of biological mechanisms, however, remains difficult. Here we introduce SQUID, a genomic DNN interpretability framework based on surrogate modeling. SQUID approximates genomic DNNs in user-specified regions of sequence space using surrogate models, i.e., simpler models that are mechanistically interpretable. Importantly, SQUID removes the confounding effects that nonlinearities and heteroscedastic noise in functional genomics data can have on model interpretation. Benchmarking analysis on multiple genomic DNNs shows that SQUID, when compared to established interpretability methods, identifies motifs that are more consistent across genomic loci and yields improved single-nucleotide variant-effect predictions. SQUID also supports surrogate models that quantify epistatic interactions within and between cis-regulatory elements. SQUID thus advances the ability to mechanistically interpret genomic DNNs.
| Item Type: | Paper |
|---|---|
| Subjects: | bioinformatics bioinformatics > genomics and proteomics organs, tissues, organelles, cell types and functions > tissues types and functions > neural networks |
| CSHL Authors: | |
| Communities: | CSHL labs > Kinney lab CSHL labs > Koo Lab CSHL labs > McCandlish lab |
| SWORD Depositor: | CSHL Elements |
| Depositing User: | CSHL Elements |
| Date: | 16 November 2023 |
| Date Deposited: | 20 Dec 2023 20:55 |
| Last Modified: | 20 Dec 2023 20:56 |
| PMCID: | PMC10680760 |
| Related URLs: | |
| URI: | https://repository.cshl.edu/id/eprint/41358 |
Actions (login required)
 |
Administrator's edit/view item |