Interpreting cis-regulatory mechanisms from genomic deep neural networks using surrogate models
Seitz, E.; McCandlish, D. M.; Kinney, J. B.; Koo, P. K.
Show 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.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
Similar papers in this journal
- Evaluating the representational power of pre-trained DNA language models for regulatory genomics 98%
- An interpretable bimodal neural network characterizes the sequence and preexisting chromatin predictors of induced TF binding 98%
- CREaTor: zero-shot cis-regulatory pattern modeling with attention mechanisms 97%
Similar papers in this journal
- Benchmarking of deep neural networks for predicting personal gene expression from DNA sequence highlights shortcomings 97%
- Sequence-based modeling of genome 3D architecture from kilobase to chromosome-scale 97%
- Dynamic network-guided CRISPRi screen reveals CTCF loop-constrained nonlinear enhancer-gene regulatory activity in cell state transitions 97%
Similar papers in this journal
- AlphaGenome: advancing regulatory variant effect prediction with a unified DNA sequence model 97%
- Massively parallel characterization of transcriptional regulatory elements in three diverse human cell types 97%
- Large-scale clinical interpretation of genetic variants using evolutionary data and deep learning 97%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.