Next Article in Journal
Comprehensive Analysis of Soybean Mosaic Virus P3 Protein Interactors and Hypersensitive Response-Like Lesion-Inducing Protein Function
Next Article in Special Issue
Targeting HIV/HCV Coinfection Using a Machine Learning-Based Multiple Quantitative Structure-Activity Relationships (Multiple QSAR) Method
Previous Article in Journal
“Biological Adhesion” is a Significantly Regulated Molecular Process during Long-Term Primary In Vitro Culture of Oviductal Epithelial Cells (Oecs): A Transcriptomic and Proteomic Study
Previous Article in Special Issue
Predicting Apoptosis Protein Subcellular Locations based on the Protein Overlapping Property Matrix and Tri-Gram Encoding
 
 
Article

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

1
Accutar Biotechnology Inc., 760 Parkside Ave., Brooklyn, NY 11226, USA
2
Amgen Inc., 1 Amgen Center Dr., Thousand Oaks, CA 91320, USA
3
Amgen Inc., 360 Binney St., Cambridge, MA 02141, USA
*
Authors to whom correspondence should be addressed.
These authors contributed equally to this work.
Int. J. Mol. Sci. 2019, 20(14), 3389; https://doi.org/10.3390/ijms20143389
Received: 22 March 2019 / Revised: 10 June 2019 / Accepted: 2 July 2019 / Published: 10 July 2019
(This article belongs to the Special Issue Artificial Intelligence and Computer Aided Drug Design)
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. View Full-Text
Keywords: deep learning; ADME prediction; drug discovery deep learning; ADME prediction; drug discovery
Show Figures

Graphical abstract

MDPI and ACS Style

Liu, K.; Sun, X.; Jia, L.; Ma, J.; Xing, H.; Wu, J.; Gao, H.; Sun, Y.; Boulnois, F.; Fan, J. Chemi-Net: A Molecular Graph Convolutional Network for Accurate Drug Property Prediction. Int. J. Mol. Sci. 2019, 20, 3389. https://doi.org/10.3390/ijms20143389

AMA Style

Liu K, Sun X, Jia L, Ma J, Xing H, Wu J, Gao H, Sun Y, Boulnois F, Fan J. Chemi-Net: A Molecular Graph Convolutional Network for Accurate Drug Property Prediction. International Journal of Molecular Sciences. 2019; 20(14):3389. https://doi.org/10.3390/ijms20143389

Chicago/Turabian Style

Liu, Ke, Xiangyan Sun, Lei Jia, Jun Ma, Haoming Xing, Junqiu Wu, Hua Gao, Yax Sun, Florian Boulnois, and Jie Fan. 2019. "Chemi-Net: A Molecular Graph Convolutional Network for Accurate Drug Property Prediction" International Journal of Molecular Sciences 20, no. 14: 3389. https://doi.org/10.3390/ijms20143389

Find Other Styles
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop