A Morse Transform for Drug Discovery

2025-03-06

AI Summary

This work addresses the low efficiency and poor interpretability of ligand binding affinity prediction in early-stage drug discovery. We propose the first topology-driven virtual screening framework grounded in piecewise-linear Morse theory. Methodologically, ligand molecules are modeled as pruned Delaunay triangulation simplicial complexes; topological features—critical points (maxima, minima, and saddle points)—are extracted via multi-directional height functions and fed into a lightweight classifier for binding potential prediction. Our key contribution is the systematic introduction of Morse theory to establish an interpretable, deep learning–free ligand representation paradigm, achieving both computational efficiency and enhanced model transparency. Evaluated on standard benchmark datasets, the framework attains state-of-the-art performance while simultaneously ensuring high predictive accuracy, intrinsic interpretability, and scalability. This advances AI-augmented rational drug design by providing a principled, topology-informed alternative to black-box deep learning approaches.

Technology Category

• Machine Learning: Other Foundations of Machine Learning
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We introduce a new ligand-based virtual screening (LBVS) framework that uses piecewise linear (PL) Morse theory to predict ligand binding potential. We model ligands as simplicial complexes via a pruned Delaunay triangulation, and catalogue the critical points across multiple directional height functions. This produces a rich feature vector, consisting of crucial topological features -- peaks, troughs, and saddles -- that characterise ligand surfaces relevant to binding interactions. Unlike contemporary LBVS methods that rely on computationally-intensive deep neural networks, we require only a lightweight classifier. The Morse theoretic approach achieves state-of-the-art performance on standard datasets while offering an interpretable feature vector and scalable method for ligand prioritization in early-stage drug discovery.

Problem

• Predict ligand binding potential using Morse theory.
• Model ligands as simplicial complexes for feature extraction.
• Provide scalable, interpretable method for drug discovery.

Innovation

• Uses piecewise linear Morse theory
• Models ligands as simplicial complexes
• Lightweight classifier for ligand prioritization