GPO-VAE: Modeling Explainable Gene Perturbation Responses utilizing GRN-Aligned Parameter Optimization

2025-01-31

AI Summary

This study addresses the problem of interpretable prediction of cellular transcriptional responses to genetic perturbations. We propose GRN-aligned VAE, a variational autoencoder that explicitly embeds prior knowledge of gene regulatory networks (GRNs) into its latent space and introduces a novel GRN-alignment parameter optimization mechanism. This mechanism enforces differentiable graph-structured regularization on perturbation-related parameters, ensuring they conform to biologically plausible regulatory logic. The method achieves state-of-the-art predictive accuracy across multiple benchmark datasets while enabling biologically interpretable modeling: the inferred GRNs exhibit strong qualitative concordance with experimentally validated pathways, and the model supports mechanistically grounded, testable interpretations of regulatory relationships. By jointly optimizing prediction fidelity and biological plausibility, GRN-aligned VAE establishes a new paradigm for mechanism-driven therapeutic intervention design.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Reasoning under Uncertainty: Graphical Models
• Computer Vision: Generative Adversarial Networks (GANs) for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Motivation: Predicting cellular responses to genetic perturbations is essential for understanding biological systems and developing targeted therapeutic strategies. While variational autoencoders (VAEs) have shown promise in modeling perturbation responses, their limited explainability poses a significant challenge, as the learned features often lack clear biological meaning. Nevertheless, model explainability is one of the most important aspects in the realm of biological AI. One of the most effective ways to achieve explainability is incorporating the concept of gene regulatory networks (GRNs) in designing deep learning models such as VAEs. GRNs elicit the underlying causal relationships between genes and are capable of explaining the transcriptional responses caused by genetic perturbation treatments. Results: We propose GPO-VAE, an explainable VAE enhanced by GRN-aligned Parameter Optimization that explicitly models gene regulatory networks in the latent space. Our key approach is to optimize the learnable parameters related to latent perturbation effects towards GRN-aligned explainability. Experimental results on perturbation prediction show our model achieves state-of-the-art performance in predicting transcriptional responses across multiple benchmark datasets. Furthermore, additional results on evaluating the GRN inference task reveal our model's ability to generate meaningful GRNs compared to other methods. According to qualitative analysis, GPO-VAE posseses the ability to construct biologically explainable GRNs that align with experimentally validated regulatory pathways. GPO-VAE is available at https://github.com/dmis-lab/GPO-VAE

Problem

• Cellular Response
• Genetic Changes
• Interpretable Modeling

Innovation

• GPO-VAE
• Gene Regulatory Networks
• Interpretability Enhancement