Causal Inference in Gene Regulatory Networks with GFlowNet: Towards Scalability in Large Systems

TL;DR

This paper introduces Swift-DynGFN, a scalable framework for causal inference in gene regulatory networks that leverages gene independence to improve efficiency and handle cyclic feedback loops.

Contribution

Swift-DynGFN is a novel method that enhances causal structure learning in GRNs by addressing scalability and cyclic dynamics through gene-wise independence exploitation.

Findings

Improves causal structure learning accuracy in GRNs.

Demonstrates scalability on large synthetic and real datasets.

Reduces computational costs via parallelization.

Abstract

Understanding causal relationships within Gene Regulatory Networks (GRNs) is essential for unraveling the gene interactions in cellular processes. However, causal discovery in GRNs is a challenging problem for multiple reasons including the existence of cyclic feedback loops and uncertainty that yields diverse possible causal structures. Previous works in this area either ignore cyclic dynamics (assume acyclic structure) or struggle with scalability. We introduce Swift-DynGFN as a novel framework that enhances causal structure learning in GRNs while addressing scalability concerns. Specifically, Swift-DynGFN exploits gene-wise independence to boost parallelization and to lower computational cost. Experiments on real single-cell RNA velocity and synthetic GRN datasets showcase the advancement in learning causal structure in GRNs and scalability in larger systems.