Chemi-Net: A Molecular Graph Convolutional Network for Accurate Drug Property Prediction.

ABSTRACT: Absorption, distribution, metabolism, and excretion (ADME) studies are critical for drug discovery. Conventionally, these tasks, together with other chemical property predictions, rely on domain-specific feature descriptors, or fingerprints. Following the recent success of neural networks, we developed Chemi-Net, a completely data-driven, domain knowledge-free, deep learning method for ADME property prediction. To compare the relative performance of Chemi-Net with Cubist, one of the popular machine learning programs used by Amgen, a large-scale ADME property prediction study was performed on-site at Amgen. For all 13 data sets, Chemi-Net resulted in higher R² values compared with the Cubist benchmark. The median R² increase rate over Cubist was 26.7%. We expect that the significantly increased accuracy of ADME prediction seen with Chemi-Net over Cubist will greatly accelerate drug discovery.

SUBMITTER: Liu K 

PROVIDER: S-EPMC6678642 | biostudies-literature | 2019 Jul

REPOSITORIES: biostudies-literature