Review of unsupervised pretraining strategies for molecules representation. (2nd August 2021)

• Record Type: Journal Article
• Authors: Yu, Linhui; Su, Yansen; Liu, Yuansheng; Zeng, Xiangxiang
• Is Part Of: Briefings in functional genomics. Volume 20:Number 5(2021)
• Page Start: 323
• Page End: 332
• Publication Date: 2021-08-02
• Subjects: drug discovery -- deep learning -- drug–drug interaction prediction -- drug–target interaction prediction -- molecular properties prediction -- unsupervised pretraining; Genomics -- Methodology -- Periodicals; Genomics -- Technological innovations -- Periodicals; 572.86072
• DOI: 10.1093/bfgp/elab036
• Languages: English
• ISSNs: 2041-2649

Abstract

Abstract: In recent years, the computer-assisted techniques make a great progress in the field of drug discovery. And, yet, the problem of limited labeled data problem is still challenging and also restricts the performance of these techniques in specific tasks, such as molecular property prediction, compound-protein interaction and de novo molecular generation. One effective solution is to utilize the experience and knowledge gained from other tasks to cope with related pursuits. Unsupervised pretraining is promising, due to its capability of leveraging a vast number of unlabeled molecules and acquiring a more informative molecular representation for the downstream tasks. In particular, models trained on large-scale unlabeled molecules can capture generalizable features, and this ability can be employed to improve the performance of specific downstream tasks. Many relevant pretraining works have been recently proposed. Here, we provide an overview of molecular unsupervised pretraining and related applications in drug discovery. Challenges and possible solutions are also summarized.