Language selection

Government of Canada / Gouvernement du Canada

Search

Geometric deep learning for protein–protein interaction predictions

From National Research Council Canada

Download
  1. (PDF, 2.0 MiB)
DOIResolve DOI: https://doi.org/10.1109/ACCESS.2022.3201543
AuthorSearch for: 1; Search for: 1ORCID identifier: https://orcid.org/0000-0001-6515-2556; Search for: ORCID identifier: https://orcid.org/0000-0003-1914-5077; Search for:
Affiliation
  1. National Research Council Canada. Digital Technologies
FunderSearch for: National Research Council Canada through the Artificial Intelligence for Design (AI4Design) Challenge Program
FormatText, Article
Subjectprotein-protein interaction; geometric deep learning; spectral convolutional neural network; graph convolutional neural network; macromolecular surface; heat kernel; wave kernel
Abstract
This work introduces novel approaches, based on geometrical deep learning, for predicting protein–protein interactions. A dataset containing both interacting and non-interacting proteins is selected from the Negatome Database. Interactions are predicted from a graph representing the proteins’ three-dimensional macromolecular surfaces. The nodes are described with heat and wave kernel signatures. Twenty-one neural network architectures are proposed and compared; these are based on graph convolutional neural networks, spectral convolutional neural networks, and a novel spatio–spectral spatialized-gated convolutional neural network. The experimental results demonstrate the accuracy and the efficiency of the proposed architectures.
Publication date
PublisherIEEE
Licence
In
  • IEEE Access 10 (25 August 2022): 90045–90055.
LanguageEnglish
Peer reviewedYes
Export citationExport as RIS
Report a correctionReport a correction (opens in a new tab)
Record identifier6615a79a-66fc-4640-881f-9aa94c806c3e
Record created2022-09-26
Record modified2025-12-18
Date modified: