Feature importance correlation from machine learning indicates functional relationships between proteins and similar compound binding characteristics.

ABSTRACT: Machine learning is widely applied in drug discovery research to predict molecular properties and aid in the identification of active compounds. Herein, we introduce a new approach that uses model-internal information from compound activity predictions to uncover relationships between target proteins. On the basis of a large-scale analysis generating and comparing machine learning models for more than 200 proteins, feature importance correlation analysis is shown to detect similar compound binding characteristics. Furthermore, rather unexpectedly, the analysis also reveals functional relationships between proteins that are independent of active compounds and binding characteristics. Feature importance correlation analysis does not depend on specific representations, algorithms, or metrics

SUBMITTER: Rodriguez-Perez R 

PROVIDER: S-EPMC8270985 | biostudies-literature | 2021 Jul

REPOSITORIES: biostudies-literature