Parametric Bayesian Priors and Better Choice of Negative Examples Improve Protein Function Prediction

Overview

Journal Bioinformatics

Publisher Oxford University Press

Specialty Biology

Date 2013 Mar 21

PMID 23511543

Citations 13

Authors

Noah Youngs

Duncan Penfold-Brown

Kevin Drew

Dennis Shasha

Richard Bonneau

Affiliations

Soon will be listed here.

Abstract

Motivation: Computational biologists have demonstrated the utility of using machine learning methods to predict protein function from an integration of multiple genome-wide data types. Yet, even the best performing function prediction algorithms rely on heuristics for important components of the algorithm, such as choosing negative examples (proteins without a given function) or determining key parameters. The improper choice of negative examples, in particular, can hamper the accuracy of protein function prediction.

Results: We present a novel approach for choosing negative examples, using a parameterizable Bayesian prior computed from all observed annotation data, which also generates priors used during function prediction. We incorporate this new method into the GeneMANIA function prediction algorithm and demonstrate improved accuracy of our algorithm over current top-performing function prediction methods on the yeast and mouse proteomes across all metrics tested.

Availability: Code and Data are available at: http://bonneaulab.bio.nyu.edu/funcprop.html

Citing Articles

Predicting protein functions using positive-unlabeled ranking with ontology-based priors.

Zhapa-Camacho F, Tang Z, Kulmanov M, Hoehndorf R Bioinformatics. 2024; 40(Suppl 1):i401-i409.

PMID: 38940168 PMC: 11211813. DOI: 10.1093/bioinformatics/btae237.

Computational Methods for Prediction of Human Protein-Phenotype Associations: A Review.

Liu L, Zhu S Phenomics. 2023; 1(4):171-185.

PMID: 36939789 PMC: 9590544. DOI: 10.1007/s43657-021-00019-w.

A Literature Review of Gene Function Prediction by Modeling Gene Ontology.

Zhao Y, Wang J, Chen J, Zhang X, Guo M, Yu G Front Genet. 2020; 11:400.

PMID: 32391061 PMC: 7193026. DOI: 10.3389/fgene.2020.00400.

Supervised learning is an accurate method for network-based gene classification.

Liu R, Mancuso C, Yannakopoulos A, Johnson K, Krishnan A Bioinformatics. 2020; 36(11):3457-3465.

PMID: 32129827 PMC: 7267831. DOI: 10.1093/bioinformatics/btaa150.

Evaluating the impact of topological protein features on the negative examples selection.

Boldi P, Frasca M, Malchiodi D BMC Bioinformatics. 2018; 19(Suppl 14):417.

PMID: 30453879 PMC: 6245585. DOI: 10.1186/s12859-018-2385-x.

References

Smoot M, Ono K, Ruscheinski J, Wang P, Ideker T . Cytoscape 2.8: new features for data integration and network visualization. Bioinformatics. 2010; 27(3):431-2. PMC: 3031041. DOI: 10.1093/bioinformatics/btq675. View

Pena-Castillo L, Tasan M, Myers C, Lee H, Joshi T, Zhang C . A critical assessment of Mus musculus gene function prediction using integrated genomic evidence. Genome Biol. 2008; 9 Suppl 1:S2. PMC: 2447536. DOI: 10.1186/gb-2008-9-s1-s2. View

Tasan M, Tian W, Hill D, Gibbons F, Blake J, Roth F . An en masse phenotype and function prediction system for Mus musculus. Genome Biol. 2008; 9 Suppl 1:S8. PMC: 2447542. DOI: 10.1186/gb-2008-9-s1-s8. View

Zhang C, Joshi T, Lin G, Xu D . An integrated probabilistic approach for gene function prediction using multiple sources of high-throughput data. Int J Comput Biol Drug Des. 2010; 1(3):254-74. DOI: 10.1504/ijcbdd.2008.021418. View

Drew K, Winters P, Butterfoss G, Berstis V, Uplinger K, Armstrong J . The Proteome Folding Project: proteome-scale prediction of structure and function. Genome Res. 2011; 21(11):1981-94. PMC: 3205581. DOI: 10.1101/gr.121475.111. View

Mostafavi S, Morris Q . Fast integration of heterogeneous data sources for predicting gene function with limited annotation. Bioinformatics. 2010; 26(14):1759-65. PMC: 2894508. DOI: 10.1093/bioinformatics/btq262. View

Gomez S, Noble W, Rzhetsky A . Learning to predict protein-protein interactions from protein sequences. Bioinformatics. 2003; 19(15):1875-81. DOI: 10.1093/bioinformatics/btg352. View

Pavlidis P, Gillis J . Progress and challenges in the computational prediction of gene function using networks. F1000Res. 2013; 1:14. PMC: 3782350. DOI: 10.12688/f1000research.1-14.v1. View

Marcotte E, Pellegrini M, Thompson M, Yeates T, Eisenberg D . A combined algorithm for genome-wide prediction of protein function. Nature. 1999; 402(6757):83-6. DOI: 10.1038/47048. View

10.

Lee H, Tu Z, Deng M, Sun F, Chen T . Diffusion kernel-based logistic regression models for protein function prediction. OMICS. 2006; 10(1):40-55. DOI: 10.1089/omi.2006.10.40. View

11.

Troyanskaya O, Dolinski K, Owen A, Altman R, Botstein D . A Bayesian framework for combining heterogeneous data sources for gene function prediction (in Saccharomyces cerevisiae). Proc Natl Acad Sci U S A. 2003; 100(14):8348-53. PMC: 166232. DOI: 10.1073/pnas.0832373100. View

12.

Obozinski G, Lanckriet G, Grant C, Jordan M, Noble W . Consistent probabilistic outputs for protein function prediction. Genome Biol. 2008; 9 Suppl 1:S6. PMC: 2447540. DOI: 10.1186/gb-2008-9-s1-s6. View

13.

Qi Y, Klein-Seetharaman J, Bar-Joseph Z . Random forest similarity for protein-protein interaction prediction from multiple sources. Pac Symp Biocomput. 2005; :531-42. View

14.

King O, Foulger R, Dwight S, White J, Roth F . Predicting gene function from patterns of annotation. Genome Res. 2003; 13(5):896-904. PMC: 430892. DOI: 10.1101/gr.440803. View

15.

Stark C, Breitkreutz B, Reguly T, Boucher L, Breitkreutz A, Tyers M . BioGRID: a general repository for interaction datasets. Nucleic Acids Res. 2005; 34(Database issue):D535-9. PMC: 1347471. DOI: 10.1093/nar/gkj109. View

16.

Tsuda K, Shin H, Scholkopf B . Fast protein classification with multiple networks. Bioinformatics. 2005; 21 Suppl 2:ii59-65. DOI: 10.1093/bioinformatics/bti1110. View

17.

Kim W, Krumpelman C, Marcotte E . Inferring mouse gene functions from genomic-scale data using a combined functional network/classification strategy. Genome Biol. 2008; 9 Suppl 1:S5. PMC: 2447539. DOI: 10.1186/gb-2008-9-s1-s5. View

18.

Leone M, Pagnani A . Predicting protein functions with message passing algorithms. Bioinformatics. 2004; 21(2):239-47. DOI: 10.1093/bioinformatics/bth491. View

19.

Huttenhower C, Hibbs M, Myers C, Caudy A, Hess D, Troyanskaya O . The impact of incomplete knowledge on evaluation: an experimental benchmark for protein function prediction. Bioinformatics. 2009; 25(18):2404-10. PMC: 2735660. DOI: 10.1093/bioinformatics/btp397. View

20.

Greene C, Troyanskaya O . Accurate evaluation and analysis of functional genomics data and methods. Ann N Y Acad Sci. 2012; 1260:95-100. PMC: 4753770. DOI: 10.1111/j.1749-6632.2011.06383.x. View