Graph Neural Networks for Predicting Protein Functions

Vassilis N. Ioannidis, Antonio G. Marques, Georgios B. Giannakis

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Learning the functions associated with a protein is essential to gaining insights for disease diagnostics, medical treatment, and human biology. In this paper, protein function prediction is posed as a semi-supervised learning task over multi-relational graphs, and it is tackled using a graph neural network (GNN) approach. The novel GNN architecture employs multi-relational graphs and weighs the influence of the different relations via learnable parameters. The ultimate goal is to design a powerful learning architecture able to: discover complex and highly nonlinear data associations, combine (and select) multiple types of relations, and scale gracefully with respect to the size of the graph. Numerical tests with protein networks corroborate the performance gains relative to state-of-the-art alternatives.

Original languageEnglish (US)
Title of host publication2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages221-225
Number of pages5
ISBN (Electronic)9781728155494
DOIs
StatePublished - Dec 2019
Event8th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2019 - Le Gosier, Guadeloupe
Duration: Dec 15 2019Dec 18 2019

Publication series

Name2019 IEEE 8th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2019 - Proceedings

Conference

Conference8th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, CAMSAP 2019
Country/TerritoryGuadeloupe
CityLe Gosier
Period12/15/1912/18/19

Keywords

  • Deep neural networks
  • graph neural networks
  • graph signals
  • multi-relational graphs
  • protein networks

Fingerprint

Dive into the research topics of 'Graph Neural Networks for Predicting Protein Functions'. Together they form a unique fingerprint.

Cite this