ScanNet: A Web Server for Structure-based Prediction of Protein Binding Sites with Geometric Deep Learning

Jérôme Tubiana; Dina Schneidman-Duhovny; Haim J. Wolfson

doi:10.1016/j.jmb.2022.167758

ScanNet: A Web Server for Structure-based Prediction of Protein Binding Sites with Geometric Deep Learning

Jérôme Tubiana^*, Dina Schneidman-Duhovny, Haim J. Wolfson

^*Corresponding author for this work

The Rachel and Selim Benin School of Engineering and Computer Science

Research output: Contribution to journal › Article › peer-review

2 Scopus citations

Abstract

Predicting the various binding sites of a protein from its structure sheds light on its function and paves the way towards design of interaction inhibitors. Here, we report ScanNet, a freely available web server for prediction of protein–protein, protein - disordered protein and protein - antibody binding sites from structure. ScanNet (Spatio-Chemical Arrangement of Neighbors Network) is an end-to-end, interpretable geometric deep learning model that learns spatio-chemical patterns directly from 3D structures. ScanNet consistently outperforms Machine Learning models based on handcrafted features and comparative modeling approaches. The web server is linked to both the PDB and AlphaFoldDB, and supports user-provided structure files. Predictions can be readily visualized on the website via the Molstar web app and locally via ChimeraX. ScanNet is available at http://bioinfo3d.cs.tau.ac.il/ScanNet/.

Original language	American English
Article number	167758
Journal	Journal of Molecular Biology
Volume	434
Issue number	19
DOIs	https://doi.org/10.1016/j.jmb.2022.167758
State	Published - 15 Oct 2022

Bibliographical note

Funding Information:
J.T. acknowledges financial support from the Edmond J. Safra Center for Bioinformatics at Tel Aviv University and from the Human Frontier Science Program (cross-disciplinary postdoctoral fellowship LT001058/2019-C). D.S. was supported by ISF 1466/18, Israel ministry of Science and Technology and HUJI-CIDR. This work was supported by Len Blavatnik and the Blavatnik Family Foundation. We are grateful to Sonia Lichtenzveig Sela and the TAU-CS system team for their technical support.

Publisher Copyright:
© 2022 Elsevier Ltd

Access to Document

10.1016/j.jmb.2022.167758

Cite this

@article{9ab954b2be454bc69c342449514d6bfb,

title = "ScanNet: A Web Server for Structure-based Prediction of Protein Binding Sites with Geometric Deep Learning",

abstract = "Predicting the various binding sites of a protein from its structure sheds light on its function and paves the way towards design of interaction inhibitors. Here, we report ScanNet, a freely available web server for prediction of protein–protein, protein - disordered protein and protein - antibody binding sites from structure. ScanNet (Spatio-Chemical Arrangement of Neighbors Network) is an end-to-end, interpretable geometric deep learning model that learns spatio-chemical patterns directly from 3D structures. ScanNet consistently outperforms Machine Learning models based on handcrafted features and comparative modeling approaches. The web server is linked to both the PDB and AlphaFoldDB, and supports user-provided structure files. Predictions can be readily visualized on the website via the Molstar web app and locally via ChimeraX. ScanNet is available at http://bioinfo3d.cs.tau.ac.il/ScanNet/.",

author = "J{\'e}r{\^o}me Tubiana and Dina Schneidman-Duhovny and Wolfson, {Haim J.}",

note = "Funding Information: J.T. acknowledges financial support from the Edmond J. Safra Center for Bioinformatics at Tel Aviv University and from the Human Frontier Science Program (cross-disciplinary postdoctoral fellowship LT001058/2019-C). D.S. was supported by ISF 1466/18, Israel ministry of Science and Technology and HUJI-CIDR. This work was supported by Len Blavatnik and the Blavatnik Family Foundation. We are grateful to Sonia Lichtenzveig Sela and the TAU-CS system team for their technical support. Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = oct,

day = "15",

doi = "10.1016/j.jmb.2022.167758",

language = "American English",

volume = "434",

journal = "Journal of Molecular Biology",

issn = "0022-2836",

publisher = "Academic Press Inc.",

number = "19",

}

TY - JOUR

T1 - ScanNet

T2 - A Web Server for Structure-based Prediction of Protein Binding Sites with Geometric Deep Learning

AU - Tubiana, Jérôme

AU - Schneidman-Duhovny, Dina

AU - Wolfson, Haim J.

N1 - Funding Information: J.T. acknowledges financial support from the Edmond J. Safra Center for Bioinformatics at Tel Aviv University and from the Human Frontier Science Program (cross-disciplinary postdoctoral fellowship LT001058/2019-C). D.S. was supported by ISF 1466/18, Israel ministry of Science and Technology and HUJI-CIDR. This work was supported by Len Blavatnik and the Blavatnik Family Foundation. We are grateful to Sonia Lichtenzveig Sela and the TAU-CS system team for their technical support. Publisher Copyright: © 2022 Elsevier Ltd

PY - 2022/10/15

Y1 - 2022/10/15

N2 - Predicting the various binding sites of a protein from its structure sheds light on its function and paves the way towards design of interaction inhibitors. Here, we report ScanNet, a freely available web server for prediction of protein–protein, protein - disordered protein and protein - antibody binding sites from structure. ScanNet (Spatio-Chemical Arrangement of Neighbors Network) is an end-to-end, interpretable geometric deep learning model that learns spatio-chemical patterns directly from 3D structures. ScanNet consistently outperforms Machine Learning models based on handcrafted features and comparative modeling approaches. The web server is linked to both the PDB and AlphaFoldDB, and supports user-provided structure files. Predictions can be readily visualized on the website via the Molstar web app and locally via ChimeraX. ScanNet is available at http://bioinfo3d.cs.tau.ac.il/ScanNet/.

AB - Predicting the various binding sites of a protein from its structure sheds light on its function and paves the way towards design of interaction inhibitors. Here, we report ScanNet, a freely available web server for prediction of protein–protein, protein - disordered protein and protein - antibody binding sites from structure. ScanNet (Spatio-Chemical Arrangement of Neighbors Network) is an end-to-end, interpretable geometric deep learning model that learns spatio-chemical patterns directly from 3D structures. ScanNet consistently outperforms Machine Learning models based on handcrafted features and comparative modeling approaches. The web server is linked to both the PDB and AlphaFoldDB, and supports user-provided structure files. Predictions can be readily visualized on the website via the Molstar web app and locally via ChimeraX. ScanNet is available at http://bioinfo3d.cs.tau.ac.il/ScanNet/.

UR - http://www.scopus.com/inward/record.url?scp=85135935580&partnerID=8YFLogxK

U2 - 10.1016/j.jmb.2022.167758

DO - 10.1016/j.jmb.2022.167758

M3 - Article

C2 - 36116806

AN - SCOPUS:85135935580

SN - 0022-2836

VL - 434

JO - Journal of Molecular Biology

JF - Journal of Molecular Biology

IS - 19

M1 - 167758

ER -

ScanNet: A Web Server for Structure-based Prediction of Protein Binding Sites with Geometric Deep Learning

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this