Representability of algebraic topology for biomolecules in machine learning based scoring and virtual screening

• Published in: PLoS computational biology Vol. 14; no. 1; p. e1005929
• Main Authors: Cang, Zixuan, Mu, Lin, Wei, Guo-Wei
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 08.01.2018; Public Library of Science (PLoS)
• Subjects: Accuracy; Algebra; Algorithms; Area Under Curve; Artificial intelligence; Artificial neural networks; Binding; Biochemistry & Molecular Biology; Bioinformatics; Biological research; Biology; Biology and Life Sciences; Biomolecules; Computational biology; Computational Biology - methods; Computational chemistry; Computer and Information Sciences; Coordination compounds; Databases, Protein; Datasets; Electrostatic properties; Electrostatics; Funding; Homology; Humans; Induction algorithms; Learning algorithms; Ligands; Machine Learning; Mathematical & Computational Biology; Mathematics; Methods; Models, Neurological; Molecular Dynamics Simulation; Molecular interactions; Neural networks; Neural Networks, Computer; Nucleic Acids - chemistry; Physical Sciences; Physics; Predictions; Protein Binding; Protein Interaction Mapping; Proteins; Proteins - chemistry; Research and Analysis Methods; Screening; Static Electricity; Topology; United States > US; East Lansing Michigan; Michigan
• DOI: 10.1371/journal.pcbi.1005929
• ISSN: 1553-7358; 1553-734X

This work introduces a number of algebraic topology approaches, including multi-component persistent homology, multi-level persistent homology, and electrostatic persistence for the representation, characterization, and description of small molecules and biomolecular complexes. In contrast to the conventional persistent homology, multi-component persistent homology retains critical chemical and biological information during the topological simplification of biomolecular geometric complexity. Multi-level persistent homology enables a tailored topological description of inter- and/or intra-molecular interactions of interest. Electrostatic persistence incorporates partial charge information into topological invariants. These topological methods are paired with Wasserstein distance to characterize similarities between molecules and are further integrated with a variety of machine learning algorithms, including k-nearest neighbors, ensemble of trees, and deep convolutional neural networks, to manifest their descriptive and predictive powers for protein-ligand binding analysis and virtual screening of small molecules. Extensive numerical experiments involving 4,414 protein-ligand complexes from the PDBBind database and 128,374 ligand-target and decoy-target pairs in the DUD database are performed to test respectively the scoring power and the discriminatory power of the proposed topological learning strategies. It is demonstrated that the present topological learning outperforms other existing methods in protein-ligand binding affinity prediction and ligand-decoy discrimination.