TY - JOUR
T1 - DeepRank
T2 - a deep learning framework for data mining 3D protein-protein interfaces
AU - Renaud, Nicolas
AU - Geng, Cunliang
AU - Georgievska, Sonja
AU - Ambrosetti, Francesco
AU - Ridder, Lars
AU - Marzella, Dario F.
AU - Réau, Manon F.
AU - Bonvin, Alexandre M.J.J.
AU - Xue, Li C.
N1 - Funding Information:
The project is supported by the ASDI grant provided by Netherlands eScience Center (Grant number ASDI.2016.043), by SURF Open Lab “Machine learning enhanced HPC applications” grant (AB/FA/10573), and by a “Computing Time on National Computer Facilities” grant (2018/ENW/00485366) from NWO (Netherlands Organization for Scientific Research). AMJJB acknowledges financial support from the European Union Horizon 2020 projects BioExcel (675728, 823830) and EOSC-hub (777536). This work was carried out on the Dutch national e-infrastructure with the support of SURF Cooperative (Grant ID: 2018/ ENW/00485366). LX and DM acknowledge financial support by the Hypatia Fellowship from Radboudumc (Rv819.52706). We thank Dr. Valeriu Codreanu, Dr. Caspar van Leeu-wen, and Dr. Damian Podareanu from SURFsara for providing HPC support for efficient data processing using Cartesius, the Dutch national supercomputer.
Publisher Copyright:
© 2021, The Author(s).
PY - 2021/12/3
Y1 - 2021/12/3
N2 - Three-dimensional (3D) structures of protein complexes provide fundamental information to decipher biological processes at the molecular scale. The vast amount of experimentally and computationally resolved protein-protein interfaces (PPIs) offers the possibility of training deep learning models to aid the predictions of their biological relevance. We present here DeepRank, a general, configurable deep learning framework for data mining PPIs using 3D convolutional neural networks (CNNs). DeepRank maps features of PPIs onto 3D grids and trains a user-specified CNN on these 3D grids. DeepRank allows for efficient training of 3D CNNs with data sets containing millions of PPIs and supports both classification and regression. We demonstrate the performance of DeepRank on two distinct challenges: The classification of biological versus crystallographic PPIs, and the ranking of docking models. For both problems DeepRank is competitive with, or outperforms, state-of-the-art methods, demonstrating the versatility of the framework for research in structural biology.
AB - Three-dimensional (3D) structures of protein complexes provide fundamental information to decipher biological processes at the molecular scale. The vast amount of experimentally and computationally resolved protein-protein interfaces (PPIs) offers the possibility of training deep learning models to aid the predictions of their biological relevance. We present here DeepRank, a general, configurable deep learning framework for data mining PPIs using 3D convolutional neural networks (CNNs). DeepRank maps features of PPIs onto 3D grids and trains a user-specified CNN on these 3D grids. DeepRank allows for efficient training of 3D CNNs with data sets containing millions of PPIs and supports both classification and regression. We demonstrate the performance of DeepRank on two distinct challenges: The classification of biological versus crystallographic PPIs, and the ranking of docking models. For both problems DeepRank is competitive with, or outperforms, state-of-the-art methods, demonstrating the versatility of the framework for research in structural biology.
UR - http://www.scopus.com/inward/record.url?scp=85120876688&partnerID=8YFLogxK
U2 - 10.1038/s41467-021-27396-0
DO - 10.1038/s41467-021-27396-0
M3 - Article
AN - SCOPUS:85120876688
SN - 2041-1723
VL - 12
SP - 1
EP - 8
JO - Nature Communications
JF - Nature Communications
IS - 1
M1 - 7068
ER -