Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations

Jia Ming Chang, Jean Francois Taly, Ionas Erb, Ting Yi Sung, Wen Lian Hsu, Chuan Yi Tang, Cedric Notredame, Emily Chia Yu Su

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Predicting protein functional classes such as localization sites and modifications plays a crucial role in function annotation. Given a tremendous amount of sequence data yielded from high-throughput sequencing experiments, the need of efficient and interpretable prediction strategies has been rapidly amplified. Our previous approach for subcellular localization prediction, PSLDoc, archives high overall accuracy for Gram-negative bacteria. However, PSLDoc is computational intensive due to incorporation of homology extension in feature extraction and probabilistic latent semantic analysis in feature reduction. Besides, prediction results generated by support vector machines are accurate but generally difficult to interpret. In this work, we incorporate three new techniques to improve efficiency and interpretability. First, homology extension is performed against a compact non-redundant database using a fast search model to reduce running time. Second, correspondence analysis (CA) is incorporated as an efficient feature reduction to generate a clear visual separation of different protein classes. Finally, functional classes are predicted by a combination of accurate compact set (CS) relation and interpretable one-nearest neighbor (1-NN) algorithm. Besides localization data sets, we also apply a human protein kinase set to validate generality of our proposed method. Experiment results demonstrate that our method make accurate prediction in a more efficient and interpretable manner. First, homology extension using a fast search on a compact database can greatly accelerate traditional running time up to twenty-five times faster without sacrificing prediction performance. This suggests that computational costs of many other predictors that also incorporate homology information can be largely reduced. In addition, CA can not only efficiently identify discriminative features but also provide a clear visualization of different functional classes. Moreover, predictions based on CS achieve 100% precision. When combined with 1-NN on unpredicted targets by CS, our method attains slightly better or comparable performance compared with the state-of-the-art systems.

Original languageEnglish
Article numbere75542
JournalPLoS One
Volume8
Issue number10
DOIs
Publication statusPublished - Oct 11 2013

Fingerprint

Running
prediction
Databases
Proteins
proteins
Gram-Negative Bacteria
Semantics
Protein Kinases
Costs and Cost Analysis
methodology
Gram-negative bacteria
protein kinases
Support vector machines
correspondence analysis
Feature extraction
Bacteria
Visualization
Experiments
Throughput
Costs

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Biochemistry, Genetics and Molecular Biology(all)
  • Medicine(all)

Cite this

Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations. / Chang, Jia Ming; Taly, Jean Francois; Erb, Ionas; Sung, Ting Yi; Hsu, Wen Lian; Tang, Chuan Yi; Notredame, Cedric; Su, Emily Chia Yu.

In: PLoS One, Vol. 8, No. 10, e75542, 11.10.2013.

Research output: Contribution to journalArticle

Chang, Jia Ming ; Taly, Jean Francois ; Erb, Ionas ; Sung, Ting Yi ; Hsu, Wen Lian ; Tang, Chuan Yi ; Notredame, Cedric ; Su, Emily Chia Yu. / Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations. In: PLoS One. 2013 ; Vol. 8, No. 10.
@article{e80b9a2c713f420591b4fe850658be78,
title = "Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations",
abstract = "Predicting protein functional classes such as localization sites and modifications plays a crucial role in function annotation. Given a tremendous amount of sequence data yielded from high-throughput sequencing experiments, the need of efficient and interpretable prediction strategies has been rapidly amplified. Our previous approach for subcellular localization prediction, PSLDoc, archives high overall accuracy for Gram-negative bacteria. However, PSLDoc is computational intensive due to incorporation of homology extension in feature extraction and probabilistic latent semantic analysis in feature reduction. Besides, prediction results generated by support vector machines are accurate but generally difficult to interpret. In this work, we incorporate three new techniques to improve efficiency and interpretability. First, homology extension is performed against a compact non-redundant database using a fast search model to reduce running time. Second, correspondence analysis (CA) is incorporated as an efficient feature reduction to generate a clear visual separation of different protein classes. Finally, functional classes are predicted by a combination of accurate compact set (CS) relation and interpretable one-nearest neighbor (1-NN) algorithm. Besides localization data sets, we also apply a human protein kinase set to validate generality of our proposed method. Experiment results demonstrate that our method make accurate prediction in a more efficient and interpretable manner. First, homology extension using a fast search on a compact database can greatly accelerate traditional running time up to twenty-five times faster without sacrificing prediction performance. This suggests that computational costs of many other predictors that also incorporate homology information can be largely reduced. In addition, CA can not only efficiently identify discriminative features but also provide a clear visualization of different functional classes. Moreover, predictions based on CS achieve 100{\%} precision. When combined with 1-NN on unpredicted targets by CS, our method attains slightly better or comparable performance compared with the state-of-the-art systems.",
author = "Chang, {Jia Ming} and Taly, {Jean Francois} and Ionas Erb and Sung, {Ting Yi} and Hsu, {Wen Lian} and Tang, {Chuan Yi} and Cedric Notredame and Su, {Emily Chia Yu}",
year = "2013",
month = "10",
day = "11",
doi = "10.1371/journal.pone.0075542",
language = "English",
volume = "8",
journal = "PLoS One",
issn = "1932-6203",
publisher = "Public Library of Science",
number = "10",

}

TY - JOUR

T1 - Efficient and Interpretable Prediction of Protein Functional Classes by Correspondence Analysis and Compact Set Relations

AU - Chang, Jia Ming

AU - Taly, Jean Francois

AU - Erb, Ionas

AU - Sung, Ting Yi

AU - Hsu, Wen Lian

AU - Tang, Chuan Yi

AU - Notredame, Cedric

AU - Su, Emily Chia Yu

PY - 2013/10/11

Y1 - 2013/10/11

N2 - Predicting protein functional classes such as localization sites and modifications plays a crucial role in function annotation. Given a tremendous amount of sequence data yielded from high-throughput sequencing experiments, the need of efficient and interpretable prediction strategies has been rapidly amplified. Our previous approach for subcellular localization prediction, PSLDoc, archives high overall accuracy for Gram-negative bacteria. However, PSLDoc is computational intensive due to incorporation of homology extension in feature extraction and probabilistic latent semantic analysis in feature reduction. Besides, prediction results generated by support vector machines are accurate but generally difficult to interpret. In this work, we incorporate three new techniques to improve efficiency and interpretability. First, homology extension is performed against a compact non-redundant database using a fast search model to reduce running time. Second, correspondence analysis (CA) is incorporated as an efficient feature reduction to generate a clear visual separation of different protein classes. Finally, functional classes are predicted by a combination of accurate compact set (CS) relation and interpretable one-nearest neighbor (1-NN) algorithm. Besides localization data sets, we also apply a human protein kinase set to validate generality of our proposed method. Experiment results demonstrate that our method make accurate prediction in a more efficient and interpretable manner. First, homology extension using a fast search on a compact database can greatly accelerate traditional running time up to twenty-five times faster without sacrificing prediction performance. This suggests that computational costs of many other predictors that also incorporate homology information can be largely reduced. In addition, CA can not only efficiently identify discriminative features but also provide a clear visualization of different functional classes. Moreover, predictions based on CS achieve 100% precision. When combined with 1-NN on unpredicted targets by CS, our method attains slightly better or comparable performance compared with the state-of-the-art systems.

AB - Predicting protein functional classes such as localization sites and modifications plays a crucial role in function annotation. Given a tremendous amount of sequence data yielded from high-throughput sequencing experiments, the need of efficient and interpretable prediction strategies has been rapidly amplified. Our previous approach for subcellular localization prediction, PSLDoc, archives high overall accuracy for Gram-negative bacteria. However, PSLDoc is computational intensive due to incorporation of homology extension in feature extraction and probabilistic latent semantic analysis in feature reduction. Besides, prediction results generated by support vector machines are accurate but generally difficult to interpret. In this work, we incorporate three new techniques to improve efficiency and interpretability. First, homology extension is performed against a compact non-redundant database using a fast search model to reduce running time. Second, correspondence analysis (CA) is incorporated as an efficient feature reduction to generate a clear visual separation of different protein classes. Finally, functional classes are predicted by a combination of accurate compact set (CS) relation and interpretable one-nearest neighbor (1-NN) algorithm. Besides localization data sets, we also apply a human protein kinase set to validate generality of our proposed method. Experiment results demonstrate that our method make accurate prediction in a more efficient and interpretable manner. First, homology extension using a fast search on a compact database can greatly accelerate traditional running time up to twenty-five times faster without sacrificing prediction performance. This suggests that computational costs of many other predictors that also incorporate homology information can be largely reduced. In addition, CA can not only efficiently identify discriminative features but also provide a clear visualization of different functional classes. Moreover, predictions based on CS achieve 100% precision. When combined with 1-NN on unpredicted targets by CS, our method attains slightly better or comparable performance compared with the state-of-the-art systems.

UR - http://www.scopus.com/inward/record.url?scp=84885404961&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84885404961&partnerID=8YFLogxK

U2 - 10.1371/journal.pone.0075542

DO - 10.1371/journal.pone.0075542

M3 - Article

C2 - 24146760

AN - SCOPUS:84885404961

VL - 8

JO - PLoS One

JF - PLoS One

SN - 1932-6203

IS - 10

M1 - e75542

ER -