Machine learning in chemoinformatics and drug discovery

• Published in: Drug discovery today 2018-08, Vol.23 (8), p.1538-1546
• Main Authors: Lo, Yu-Chen, Rensi, Stefano E., Torng, Wen, Altman, Russ B.
• Format: Article
• Language: English
• Subjects: Animals; Diffusion of Innovation; Drug Discovery - methods; High-Throughput Screening Assays; Humans; Informatics; Machine Learning; Molecular Structure; Pattern Recognition, Automated; Pharmaceutical Preparations - chemistry; Quantitative Structure-Activity Relationship
• ISSN: 1359-6446; 1878-5832
• DOI: 10.1016/j.drudis.2018.05.010

•Chemical graph theory and descriptors in drug discovery.•Chemical fingerprint and similarity analysis.•Machine learning models for virtual screening.•Future challenges and direction in machine-learning-based drug discovery. Chemoinformatics is an established discipline focusing on extracting, processing and extrapolating meaningful data from chemical structures. With the rapid explosion of chemical ‘big’ data from HTS and combinatorial synthesis, machine learning has become an indispensable tool for drug designers to mine chemical information from large compound databases to design drugs with important biological properties. To process the chemical data, we first reviewed multiple processing layers in the chemoinformatics pipeline followed by the introduction of commonly used machine learning models in drug discovery and QSAR analysis. Here, we present basic principles and recent case studies to demonstrate the utility of machine learning techniques in chemoinformatics analyses; and we discuss limitations and future directions to guide further development in this evolving field.