Learning to Discover Regulatory Elements for Gene Expression Prediction

2025-02-19

AI Summary

This study addresses the end-to-end prediction of gene expression from DNA sequences, aiming to automatically identify key regulatory elements driving expression. We propose a causally aware information bottleneck framework that models information compression via the Beta distribution, jointly decomposes epigenetic signals and DNA sequence inputs, and explicitly characterizes the causal activity of regulatory elements—thereby effectively filtering non-causal noise. Integrating deep sequence modeling with an interpretable localization mechanism, our method achieves precise identification of regulatory regions. On multiple benchmark datasets, our approach significantly outperforms state-of-the-art baselines in gene expression prediction accuracy. Moreover, the identified regulatory regions exhibit superior biological consistency and functional enrichment compared to those detected by statistical peak-calling methods (e.g., MACS3). Our work establishes an interpretable, causally grounded computational paradigm for deciphering non-coding genomic function.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

We consider the problem of predicting gene expressions from DNA sequences. A key challenge of this task is to find the regulatory elements that control gene expressions. Here, we introduce Seq2Exp, a Sequence to Expression network explicitly designed to discover and extract regulatory elements that drive target gene expression, enhancing the accuracy of the gene expression prediction. Our approach captures the causal relationship between epigenomic signals, DNA sequences and their associated regulatory elements. Specifically, we propose to decompose the epigenomic signals and the DNA sequence conditioned on the causal active regulatory elements, and apply an information bottleneck with the Beta distribution to combine their effects while filtering out non-causal components. Our experiments demonstrate that Seq2Exp outperforms existing baselines in gene expression prediction tasks and discovers influential regions compared to commonly used statistical methods for peak detection such as MACS3. The source code is released as part of the AIRS library (https://github.com/divelab/AIRS/).

Problem

• Predicting gene expressions from DNA sequences
• Discovering regulatory elements controlling gene expressions
• Enhancing accuracy with Seq2Exp network

Innovation

• Seq2Exp network
• causal regulatory elements
• information bottleneck Beta