Probabilistic methods for predicting protein functions in protein-protein interaction networks

Christoph Best; Ralf Zimmer; Joannis Apostolakis

Probabilistic methods for predicting protein functions in protein-protein interaction networks

Christoph Best^*
, Ralf Zimmer
, Joannis Apostolakis

^*Corresponding author for this work

Ludwig Maximilian University of Munich

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic methods and of a standard support vector machine. It turns out that, with the currently available networks, the simple methods based on counting frequencies perform as well as the more sophisticated approaches.

Original language	English
Title of host publication	Proceedings of the German Conference on Bioinformatics, GCB 2004
Editors	Robert Giegerich, Jens Stoye
Publisher	Gesellschaft fur Informatik (GI)
Pages	159-168
Number of pages	10
ISBN (Electronic)	3885793822
State	Published - 2004
Externally published	Yes
Event	2004 German Conference on Bioinformatics, GCB 2004 - Bielefeld, Germany Duration: 4 Oct 2004 → 6 Oct 2004

Publication series

Name	Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)
Volume	P-53
ISSN (Print)	1617-5468
ISSN (Electronic)	2944-7682

Conference

Conference	2004 German Conference on Bioinformatics, GCB 2004
Country/Territory	Germany
City	Bielefeld
Period	4/10/04 → 6/10/04

Cite this

Best, C., Zimmer, R., & Apostolakis, J. (2004). Probabilistic methods for predicting protein functions in protein-protein interaction networks. In R. Giegerich, & J. Stoye (Eds.), Proceedings of the German Conference on Bioinformatics, GCB 2004 (pp. 159-168). (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. P-53). Gesellschaft fur Informatik (GI).

Best, Christoph ; Zimmer, Ralf ; Apostolakis, Joannis. / Probabilistic methods for predicting protein functions in protein-protein interaction networks. Proceedings of the German Conference on Bioinformatics, GCB 2004. editor / Robert Giegerich ; Jens Stoye. Gesellschaft fur Informatik (GI), 2004. pp. 159-168 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

@inproceedings{a81b7c43163843ccba8334e52c97cc47,

title = "Probabilistic methods for predicting protein functions in protein-protein interaction networks",

abstract = "We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic methods and of a standard support vector machine. It turns out that, with the currently available networks, the simple methods based on counting frequencies perform as well as the more sophisticated approaches.",

author = "Christoph Best and Ralf Zimmer and Joannis Apostolakis",

note = "Publisher Copyright: {\textcopyright} 2004 Gesellschaft fur Informatik (GI). All rights reserved.; 2004 German Conference on Bioinformatics, GCB 2004 ; Conference date: 04-10-2004 Through 06-10-2004",

year = "2004",

language = "English",

series = "Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)",

publisher = "Gesellschaft fur Informatik (GI)",

pages = "159--168",

editor = "Robert Giegerich and Jens Stoye",

booktitle = "Proceedings of the German Conference on Bioinformatics, GCB 2004",

}

Best, C, Zimmer, R & Apostolakis, J 2004, Probabilistic methods for predicting protein functions in protein-protein interaction networks. in R Giegerich & J Stoye (eds), Proceedings of the German Conference on Bioinformatics, GCB 2004. Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI), vol. P-53, Gesellschaft fur Informatik (GI), pp. 159-168, 2004 German Conference on Bioinformatics, GCB 2004, Bielefeld, Germany, 4/10/04.

Probabilistic methods for predicting protein functions in protein-protein interaction networks. / Best, Christoph; Zimmer, Ralf; Apostolakis, Joannis.
Proceedings of the German Conference on Bioinformatics, GCB 2004. ed. / Robert Giegerich; Jens Stoye. Gesellschaft fur Informatik (GI), 2004. p. 159-168 (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI); Vol. P-53).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Probabilistic methods for predicting protein functions in protein-protein interaction networks

AU - Best, Christoph

AU - Zimmer, Ralf

AU - Apostolakis, Joannis

PY - 2004

Y1 - 2004

N2 - We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic methods and of a standard support vector machine. It turns out that, with the currently available networks, the simple methods based on counting frequencies perform as well as the more sophisticated approaches.

AB - We discuss probabilistic methods for predicting protein functions from protein-protein interaction networks. Previous work based on Markov Randon Fields is extended and compared to a general machine-learning theoretic approach. Using actual protein interaction networks for yeast from the MIPS database and GO-SLIM function assignments, we compare the predictions of the different probabilistic methods and of a standard support vector machine. It turns out that, with the currently available networks, the simple methods based on counting frequencies perform as well as the more sophisticated approaches.

UR - https://www.scopus.com/pages/publications/84925053964

M3 - Conference contribution

AN - SCOPUS:84925053964

T3 - Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)

SP - 159

EP - 168

BT - Proceedings of the German Conference on Bioinformatics, GCB 2004

A2 - Giegerich, Robert

A2 - Stoye, Jens

PB - Gesellschaft fur Informatik (GI)

T2 - 2004 German Conference on Bioinformatics, GCB 2004

Y2 - 4 October 2004 through 6 October 2004

ER -

Best C, Zimmer R, Apostolakis J. Probabilistic methods for predicting protein functions in protein-protein interaction networks. In Giegerich R, Stoye J, editors, Proceedings of the German Conference on Bioinformatics, GCB 2004. Gesellschaft fur Informatik (GI). 2004. p. 159-168. (Lecture Notes in Informatics (LNI), Proceedings - Series of the Gesellschaft fur Informatik (GI)).

Probabilistic methods for predicting protein functions in protein-protein interaction networks

Abstract

Publication series

Conference

Other files and links

Fingerprint

Cite this