TY - JOUR
T1 - Guiding conventional protein-ligand docking software with convolutional neural networks
AU - Jiang, Huaipan
AU - Fan, Mengran
AU - Wang, Jian
AU - Sarma, Anup
AU - Mohanty, Shruti
AU - Dokholyan, Nikolay V.
AU - Mahdavi, Mehrdad
AU - Kandemir, Mahmut T.
N1 - Funding Information:
We acknowledge the support from the National Institutes for Health and the National Science Foundation. N.V.D. and J.W. are supported by National Institutes for Health (2R01 GM114015 and 1R35 GM134864 to N.V.D.) and the Passan Foundation. They are also supported by the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1 TR002014. M.T.K. is supported by NSF grants 1763681, 1629915, 1629129, 1931531, 0821527, 2040667, and 1439057. H.J. and M.F. are also supported by NSF grant 1439057. A.S. is supported by NSF grant 1955815. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or NSF.
Publisher Copyright:
© 2020 American Chemical Society.
PY - 2020/10/26
Y1 - 2020/10/26
N2 - The high-performance computational techniques have brought significant benefits for drug discovery efforts in recent decades. One of the most challenging problems in drug discovery is the protein-ligand binding pose prediction. To predict the most stable structure of the complex, the performance of conventional structure-based molecular docking methods heavily depends on the accuracy of scoring or energy functions (as an approximation of affinity) for each pose of the protein-ligand docking complex to effectively guide the search in an exponentially large solution space. However, due to the heterogeneity of molecular structures, the existing scoring calculation methods are either tailored to a particular data set or fail to exhibit high accuracy. In this paper, we propose a convolutional neural network (CNN)-based model that learns to predict the stability factor of the protein-ligand complex and exhibits the ability of CNNs to improve the existing docking software. Evaluated results on PDBbind data set indicate that our approach reduces the execution time of the traditional docking-based method while improving the accuracy. Our code, experiment scripts, and pretrained models are available at https://github. com/j9650/MedusaNet.
AB - The high-performance computational techniques have brought significant benefits for drug discovery efforts in recent decades. One of the most challenging problems in drug discovery is the protein-ligand binding pose prediction. To predict the most stable structure of the complex, the performance of conventional structure-based molecular docking methods heavily depends on the accuracy of scoring or energy functions (as an approximation of affinity) for each pose of the protein-ligand docking complex to effectively guide the search in an exponentially large solution space. However, due to the heterogeneity of molecular structures, the existing scoring calculation methods are either tailored to a particular data set or fail to exhibit high accuracy. In this paper, we propose a convolutional neural network (CNN)-based model that learns to predict the stability factor of the protein-ligand complex and exhibits the ability of CNNs to improve the existing docking software. Evaluated results on PDBbind data set indicate that our approach reduces the execution time of the traditional docking-based method while improving the accuracy. Our code, experiment scripts, and pretrained models are available at https://github. com/j9650/MedusaNet.
UR - http://www.scopus.com/inward/record.url?scp=85094678066&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85094678066&partnerID=8YFLogxK
U2 - 10.1021/acs.jcim.0c00542
DO - 10.1021/acs.jcim.0c00542
M3 - Article
C2 - 33100014
AN - SCOPUS:85094678066
VL - 60
SP - 4594
EP - 4602
JO - Journal of Chemical Information and Modeling
JF - Journal of Chemical Information and Modeling
SN - 1549-9596
IS - 10
ER -