Precise modelling and interpretation of bioactivities of ligands targeting G protein-coupled receptors

• DOI: 10.1093/bioinformatics/btz336
• Publication Date: 2019-06-11T19:11:41Z
• Authors: Jiansheng Wu; Ben Liu; Wallace K B Chan; Weijian Wu; Tao Pang; Haifeng Hu; Shancheng Yan; Xiaoyan Ke; Yang Zhang
• Keywords: Lasso

Accurate prediction and interpretation of ligand bioactivities are essential for virtual screening drug discovery. Unfortunately, many important targets lack experimental data about the bioactivities; this is particularly true G protein-coupled receptors (GPCRs), which account a third drugs currently on market. Computational approaches with potential precise assessment determination key substructural features determine needed to address issue.A new method, SED, was proposed predict recognize substructures associated GPCRs through coupling Lasso long extended-connectivity fingerprints (ECFPs) deep neural network training. The SED pipeline contains three successive steps: (i) representation ECFPs molecules, (ii) feature selection by (iii) bioactivity regression model. method examined set 16 representative that cover most subfamilies human GPCRs, where each has 300-5000 associations. results show achieves excellent performance in modelling bioactivities, especially those GPCR datasets without sufficient associations, improved baseline predictors 12% correlation coefficient (r2) 19% root mean square error. Detail analyses suggest major advantage lies its ability detect from significantly improves predictive performance.The source code freely available at https://zhanglab.ccmb.med.umich.edu/SED/.Supplementary Bioinformatics online.