Kolmogorov-Arnold Network for Gene Regulatory Network Inference

2025-06-16

AI Summary

Inferring gene regulatory networks (GRNs) from single-cell RNA sequencing (scRNA-seq) data faces two key challenges: (i) prevailing tree-based methods (e.g., GENIE3, GRNBOOST2) cannot distinguish activating from repressive regulatory interactions, and (ii) they lack the capacity to model continuous cellular dynamics. Method: This work introduces, for the first time, differentiable Kolmogorov–Arnold Networks (KANs) into GRN inference—integrating interpretable AI with geometric analysis to automatically infer regulatory directionality without requiring prior graph structure, while modeling smooth biological dynamics via differentiable functions. Contribution/Results: Evaluated on the BEELINE benchmark, our approach achieves absolute improvements of 5.40–28.37% in AUROC and 1.97–40.45% in AUPRC over state-of-the-art signed GRN models, demonstrating substantial gains in both sensitivity and precision for directional GRN reconstruction.

Technology Category

• Reasoning under Uncertainty: Graphical Models
• Machine Learning: Kernel Methods
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Gene regulation is central to understanding cellular processes and development, potentially leading to the discovery of new treatments for diseases and personalized medicine. Inferring gene regulatory networks (GRNs) from single-cell RNA sequencing (scRNA-seq) data presents significant challenges due to its high dimensionality and complexity. Existing tree-based models, such as GENIE3 and GRNBOOST2, demonstrated scalability and explainability in GRN inference, but they cannot distinguish regulation types nor effectively capture continuous cellular dynamics. In this paper, we introduce scKAN, a novel model that employs a Kolmogorov-Arnold network (KAN) with explainable AI to infer GRNs from scRNA-seq data. By modeling gene expression as differentiable functions matching the smooth nature of cellular dynamics, scKAN can accurately and precisely detect activation and inhibition regulations through explainable AI and geometric tools. We conducted extensive experiments on the BEELINE benchmark, and scKAN surpasses and improves the leading signed GRN inference models ranging from 5.40% to 28.37% in AUROC and from 1.97% to 40.45% in AUPRC. These results highlight the potential of scKAN in capturing the underlying biological processes in gene regulation without prior knowledge of the graph structure.

Problem

• Infer gene regulatory networks from scRNA-seq data
• Distinguish regulation types in cellular dynamics
• Improve accuracy in detecting activation and inhibition

Innovation

• Uses Kolmogorov-Arnold Network for GRN inference
• Incorporates explainable AI and geometric tools
• Models gene expression as differentiable functions