A chronopharmacology-friendly multi-target therapeutics based on AI: the example of therapeutic hypothermia

Abstract

AbstractNowadays, the complexity of disease mechanisms and inadequacy of single-target therapies in restoring the biological system has inevitably instigated the strategy of multi-target therapeutics with the application of hybrid and chimeric drugs. However, the related method is still unable to solve the conflicts between targets or between drugs. With the release of high-precision protein structure prediction artificial intelligence (AI), large-scale high-precision protein structure prediction and docking become possible. In this article, we propose a multi-target drug discovery method. Then we take an example of therapeutic hypothermia (TH). We performed protein structure prediction for all targets of each group by AlphaFold2 and RoseTTAFold. QuickVina 2 is then used for molecular docking of the proteins and drugs. After docking, we use PageRank to get the rank of drugs and drug combinations of each group. Given the differences in the scoring of different proteins, the method can effectively avoid inhibiting beneficial proteins. So it’s friendly to chronopharmacology. This method also have potential in precision medicine for its high compatibility with bioinformatics.