Mathematical deep learning for pose and binding affinity prediction and ranking in D3R Grand Challenges

• Published in: Journal of computer-aided molecular design Vol. 33; no. 1; pp. 71 - 82
• Main Authors: Nguyen, Duc Duy, Cang, Zixuan, Wu, Kedi, Wang, Menglun, Cao, Yin, Wei, Guo-Wei
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.01.2019; Springer Nature B.V
• Subjects: Affinity; Animal Anatomy; Artificial neural networks; Binding; Binding Sites; CAD; Cathepsins - chemistry; Chemistry; Chemistry and Materials Science; Computer aided design; Computer Applications in Chemistry; Crystallography, X-Ray; Databases, Protein; Deep Learning; Drug Design; Free energy; Histology; Homology; Ions; Ligands; Machine learning; Mathematical analysis; Mathematical models; Molecular Docking Simulation - methods; Morphology; Physical Chemistry; Protein Binding; Protein Conformation; Protein Kinases - chemistry; Ranking; Receptors, Cytoplasmic and Nuclear - chemistry; Thermodynamics; Drug design; Graph theory; Binding affinity; Pose prediction; Algebraic topology; Machine learning
• ISSN: 0920-654X; 1573-4951
• DOI: 10.1007/s10822-018-0146-6

Advanced mathematics, such as multiscale weighted colored subgraph and element specific persistent homology, and machine learning including deep neural networks were integrated to construct mathematical deep learning models for pose and binding affinity prediction and ranking in the last two D3R Grand Challenges in computer-aided drug design and discovery. D3R Grand Challenge 2 focused on the pose prediction, binding affinity ranking and free energy prediction for Farnesoid X receptor ligands. Our models obtained the top place in absolute free energy prediction for free energy set 1 in stage 2. The latest competition, D3R Grand Challenge 3 (GC3), is considered as the most difficult challenge so far. It has five subchallenges involving Cathepsin S and five other kinase targets, namely VEGFR2, JAK2, p38- α , TIE2, and ABL1. There is a total of 26 official competitive tasks for GC3. Our predictions were ranked 1st in 10 out of these 26 tasks.