Predicting the Absorption Potential of Chemical Compounds Through a Deep Learning Approach

• Published in: IEEE/ACM transactions on computational biology and bioinformatics Vol. 15; no. 2; pp. 432 - 440
• Main Authors: Shin, Moonshik, Jang, Donjin, Nam, Hojung, Lee, Kwang Hyung, Lee, Doheon
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.03.2018
• Subjects: Absorption; absorption prediction; Absorption, Physicochemical - drug effects; Caco-2 Cells; Caco-2 permeability; Cell Membrane Permeability - drug effects; Chemical compounds; Chemicals; Compounds; Computational Biology; Computer Simulation; Deep Learning; Drug Evaluation, Preclinical; Drugs; Humans; Machine learning; Models, Statistical; neural nets; Permeability; Pharmaceutical Preparations - metabolism; Predictive models
• ISSN: 1545-5963; 1557-9964
• DOI: 10.1109/TCBB.2016.2535233

The human colorectal carcinoma cell line (Caco-2) is a commonly used in-vitro test that predicts the absorption potential of orally administered drugs. In-silico prediction methods, based on the Caco-2 assay data, may increase the effectiveness of the high-throughput screening of new drug candidates. However, previously developed in-silico models that predict the Caco-2 cellular permeability of chemical compounds use handcrafted features that may be dataset-specific and induce over-fitting problems. Deep Neural Network (DNN) generates high-level features based on non-linear transformations for raw features, which provides high discriminant power and, therefore, creates a good generalized model. We present a DNN-based binary Caco-2 permeability classifier. Our model was constructed based on 663 chemical compounds with in-vitro Caco-2 apparent permeability data. Two hundred nine molecular descriptors are used for generating the high-level features during DNN model generation. Dropout regularization is applied to solve the over-fitting problem and the non-linear activation. The Rectified Linear Unit (ReLU) is adopted to reduce the vanishing gradient problem. The results demonstrate that the high-level features generated by the DNN are more robust than handcrafted features for predicting the cellular permeability of structurally diverse chemical compounds in Caco-2 cell lines.