Institute Homepage
Institute Homepage DE Sign In

Back

Empirical Inference Article 2009

Protein Functional Class Prediction With a Combined Graph

Thumb ticker sm l1170153
Empirical Inference
Bernhard Schölkopf
  • Director
no image
Empirical Inference
Koji Tsuda

In bioinformatics, there exist multiple descriptions of graphs for the same set of genes or proteins. For instance, in yeast systems, graph edges can represent different relationships such as protein–protein interactions, genetic interactions, or co-participation in a protein complex, etc. Relying on similarities between nodes, each graph can be used independently for prediction of protein function. However, since different graphs contain partly independent and partly complementary information about the problem at hand, one can enhance the total information extracted by combining all graphs. In this paper, we propose a method for integrating multiple graphs within a framework of semi-supervised learning. The method alternates between minimizing the objective function with respect to network output and with respect to combining weights. We apply the method to the task of protein functional class prediction in yeast. The proposed method performs significantly better than the same algorithm trained on any singl e graph.

Author(s): Shin, HH. and Tsuda, K. and Schölkopf, B.
Journal: Expert Systems with Applications
Volume: 36
Number (issue): 2
Pages: 3284-3292
Year: 2009
Month: March
Day: 0
Bibtex Type: Article (article)
DOI: 10.1016/j.eswa.2008.01.006
Digital: 0
Electronic Archiving: grant_archive
Language: en
Organization: Max-Planck-Gesellschaft
School: Biologische Kybernetik
Links:

BibTex 

@article{5086,
  title = {Protein Functional Class Prediction With a Combined Graph},
  journal = {Expert Systems with Applications},
  abstract = {In bioinformatics, there exist multiple descriptions of graphs for the same set of genes or proteins. For instance, in yeast systems, graph edges can represent different relationships such as protein–protein interactions, genetic interactions, or co-participation in a protein complex, etc. Relying on similarities between nodes, each graph can be used independently for prediction of protein function. However, since different graphs contain partly independent and partly complementary information about the problem at hand, one can enhance the total information extracted by combining all graphs. In this paper, we propose a method for integrating multiple graphs within a framework of semi-supervised learning. The method alternates between minimizing the objective function with respect to network output and with respect to combining weights. We apply the method to the task of protein functional class prediction in yeast. The proposed method performs significantly better than the same algorithm trained on any singl
  e graph.},
  volume = {36},
  number = {2},
  pages = {3284-3292},
  organization = {Max-Planck-Gesellschaft},
  school = {Biologische Kybernetik},
  month = mar,
  year = {2009},
  slug = {5086},
  author = {Shin, HH. and Tsuda, K. and Sch{\"o}lkopf, B.},
  month_numeric = {3}
}