PNNL

Abstract

Recently, Suwayyid and Wei have introduced commutative algebra as an emerging paradigm for machine learning and data science. In this work, we propose commutative algebra machine learning (CAML) for the prediction of protein-ligand binding affinities. Specifically, we apply persistent Stanley–Reisner theory, a key concept in combinatorial commutative algebra, to the affinity predictions of protein-ligand binding and metalloprotein-ligand binding. We present three new algorithms, i.e., elementspecific commutative algebra, category-specific commutative algebra, and commutative algebra on bipartite complexes, to tackle the complexity of data involved in (metallo) protein-ligand complexes. We show that the proposed CAML outperforms other stateof-the-art methods in (metallo) protein-ligand binding affinity predictions, indicating the great potential of commutative algebra learning.