Identifying protein complexes with fuzzy machine learning model

Bo Xu, Hongfei Lin, Kavishwar B. Wagholikar, Zhihao Yang, Hongfang D Liu

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Background: Many computational approaches have been developed to detect protein complexes from proteinprotein interaction (PPI) networks. However, these PPI networks are always built from high-throughput experiments. The presence of unreliable interactions in PPI network makes this task very challenging. Methods: In this study, we proposed a Genetic-Algorithm Fuzzy Naïve Bayes (GAFNB) filter to classify the protein complexes from candidate subgraphs. It takes unreliability into consideration and tackles the presence of unreliable interactions in protein complex. We first got candidate protein complexes through existed popular methods. Each candidate protein complex is represented by 29 graph features and 266 biological property based features. GAFNB model is then applied to classify the candidate complexes into positive or negative. Results: Our evaluation indicates that the protein complex identification algorithms using the GAFNB model filtering outperform original ones. For evaluation of GAFNB model, we also compared the performance of GAFNB with Naïve Bayes (NB). Results show that GAFNB performed better than NB. It indicates that a fuzzy model is more suitable when unreliability is present. Conclusions: We conclude that filtering candidate protein complexes with GAFNB model can improve the effectiveness of protein complex identification. It is necessary to consider the unreliability in this task.

Original languageEnglish (US)
Article numberS21
JournalProteome Science
Volume11
DOIs
StatePublished - Jan 1 2013

Fingerprint

Learning systems
Genetic algorithms
Proteins
Identification (control systems)
Machine Learning
Throughput
Experiments

ASJC Scopus subject areas

  • Biochemistry
  • Molecular Biology

Cite this

Identifying protein complexes with fuzzy machine learning model. / Xu, Bo; Lin, Hongfei; Wagholikar, Kavishwar B.; Yang, Zhihao; Liu, Hongfang D.

In: Proteome Science, Vol. 11, S21, 01.01.2013.

Research output: Contribution to journalArticle

Xu, Bo ; Lin, Hongfei ; Wagholikar, Kavishwar B. ; Yang, Zhihao ; Liu, Hongfang D. / Identifying protein complexes with fuzzy machine learning model. In: Proteome Science. 2013 ; Vol. 11.
@article{403bce34bd2a49be911b3ec60c3550c9,
title = "Identifying protein complexes with fuzzy machine learning model",
abstract = "Background: Many computational approaches have been developed to detect protein complexes from proteinprotein interaction (PPI) networks. However, these PPI networks are always built from high-throughput experiments. The presence of unreliable interactions in PPI network makes this task very challenging. Methods: In this study, we proposed a Genetic-Algorithm Fuzzy Na{\"i}ve Bayes (GAFNB) filter to classify the protein complexes from candidate subgraphs. It takes unreliability into consideration and tackles the presence of unreliable interactions in protein complex. We first got candidate protein complexes through existed popular methods. Each candidate protein complex is represented by 29 graph features and 266 biological property based features. GAFNB model is then applied to classify the candidate complexes into positive or negative. Results: Our evaluation indicates that the protein complex identification algorithms using the GAFNB model filtering outperform original ones. For evaluation of GAFNB model, we also compared the performance of GAFNB with Na{\"i}ve Bayes (NB). Results show that GAFNB performed better than NB. It indicates that a fuzzy model is more suitable when unreliability is present. Conclusions: We conclude that filtering candidate protein complexes with GAFNB model can improve the effectiveness of protein complex identification. It is necessary to consider the unreliability in this task.",
author = "Bo Xu and Hongfei Lin and Wagholikar, {Kavishwar B.} and Zhihao Yang and Liu, {Hongfang D}",
year = "2013",
month = "1",
day = "1",
doi = "10.1186/1477-5956-11-S1-S21",
language = "English (US)",
volume = "11",
journal = "Proteome Science",
issn = "1477-5956",
publisher = "BioMed Central",

}

TY - JOUR

T1 - Identifying protein complexes with fuzzy machine learning model

AU - Xu, Bo

AU - Lin, Hongfei

AU - Wagholikar, Kavishwar B.

AU - Yang, Zhihao

AU - Liu, Hongfang D

PY - 2013/1/1

Y1 - 2013/1/1

N2 - Background: Many computational approaches have been developed to detect protein complexes from proteinprotein interaction (PPI) networks. However, these PPI networks are always built from high-throughput experiments. The presence of unreliable interactions in PPI network makes this task very challenging. Methods: In this study, we proposed a Genetic-Algorithm Fuzzy Naïve Bayes (GAFNB) filter to classify the protein complexes from candidate subgraphs. It takes unreliability into consideration and tackles the presence of unreliable interactions in protein complex. We first got candidate protein complexes through existed popular methods. Each candidate protein complex is represented by 29 graph features and 266 biological property based features. GAFNB model is then applied to classify the candidate complexes into positive or negative. Results: Our evaluation indicates that the protein complex identification algorithms using the GAFNB model filtering outperform original ones. For evaluation of GAFNB model, we also compared the performance of GAFNB with Naïve Bayes (NB). Results show that GAFNB performed better than NB. It indicates that a fuzzy model is more suitable when unreliability is present. Conclusions: We conclude that filtering candidate protein complexes with GAFNB model can improve the effectiveness of protein complex identification. It is necessary to consider the unreliability in this task.

AB - Background: Many computational approaches have been developed to detect protein complexes from proteinprotein interaction (PPI) networks. However, these PPI networks are always built from high-throughput experiments. The presence of unreliable interactions in PPI network makes this task very challenging. Methods: In this study, we proposed a Genetic-Algorithm Fuzzy Naïve Bayes (GAFNB) filter to classify the protein complexes from candidate subgraphs. It takes unreliability into consideration and tackles the presence of unreliable interactions in protein complex. We first got candidate protein complexes through existed popular methods. Each candidate protein complex is represented by 29 graph features and 266 biological property based features. GAFNB model is then applied to classify the candidate complexes into positive or negative. Results: Our evaluation indicates that the protein complex identification algorithms using the GAFNB model filtering outperform original ones. For evaluation of GAFNB model, we also compared the performance of GAFNB with Naïve Bayes (NB). Results show that GAFNB performed better than NB. It indicates that a fuzzy model is more suitable when unreliability is present. Conclusions: We conclude that filtering candidate protein complexes with GAFNB model can improve the effectiveness of protein complex identification. It is necessary to consider the unreliability in this task.

UR - http://www.scopus.com/inward/record.url?scp=85039981152&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039981152&partnerID=8YFLogxK

U2 - 10.1186/1477-5956-11-S1-S21

DO - 10.1186/1477-5956-11-S1-S21

M3 - Article

AN - SCOPUS:85039981152

VL - 11

JO - Proteome Science

JF - Proteome Science

SN - 1477-5956

M1 - S21

ER -