Data-Driven Symbolic Higher-Order Epistasis Discovery with Kolmogorov-Arnold Networks
Oankar R. Patil, Kamran Shazand, Benoit Marteau, Yuxuan Shen, May D. Wang
This paper introduces EPIK, a deep learning framework for detecting higher-order epistasis in GWAS data, showing improved performance on simulated and real datasets and identifying new gene interactions in Parkinson's Disease.
ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics (ACM-BCB) · Oct 12, 2025Peer-reviewedGenomics