Supervised approaches for function prediction of proteins contact networks from topological structure information

Alessio Martino, Enrico Maiorino, Alessandro Giuliani, Mauro Giampieri, Antonello Rizzi

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


The role performed by a protein is directly connected to its physico-chemical structure. How the latter affects the behaviour of these molecules is still an open research topic. In this paper we consider a subset of the Escherichia Coli proteome where each protein is represented through the spectral characteristics of its residue contact network and its physiological function is encoded by a suitable class label. By casting this problem as a machine learning task, we aim at assessing whether a relation exists between such spectral properties and the protein’s function. To this end we adopted a set of supervised learning techniques, possibly optimised by means of genetic algorithms. First results are promising and they show that such high-level spectral representation contains enough information in order to discriminate among functional classes. Our experiments pave the way for further research and analysis.

Original languageEnglish
Title of host publicationImage Analysis - 20th Scandinavian Conference, SCIA 2017, Proceedings
PublisherSpringer Verlag
Number of pages12
ISBN (Print)9783319591254
Publication statusPublished - Jan 1 2017
Event20th Scandinavian Conference on Image Analysis, SCIA 2017 - Tromso, Norway
Duration: Jun 12 2017Jun 14 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10269 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference20th Scandinavian Conference on Image Analysis, SCIA 2017


  • Normalised laplacian matrix
  • Pattern recognition
  • Protein contact networks
  • Supervised learning
  • Support vector machines

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)


Dive into the research topics of 'Supervised approaches for function prediction of proteins contact networks from topological structure information'. Together they form a unique fingerprint.

Cite this