Disease-causing pathogens such as viruses introduce their proteins into the host cells in which they interact with the host's proteins, enabling the virus to replicate inside the host. These interactions between pathogen and host proteins are key to understanding infectious diseases. Often multiple diseases involve phylogenetically related or biologically similar pathogens. Here we present a multitask learning method to jointly model interactions between human proteins and three different but related viruses: Hepatitis C, Ebola virus, and Influenza A. Our multitask matrix completion-based model uses a shared low-rank structure in addition to a task-specific sparse structure to incorporate the various interactions. We obtain between 7 and 39 percentage points improvement in predictive performance over prior state-of-the-art models. We show how our model's parameters can be interpreted to reveal both general and specific interaction-relevant characteristics of the viruses. Our code is available online.*
Get full access to this article
View all access options for this article.
References
1.
AbernethyJ., BachF., EvgeniouT., and VertJ.P. (2009). A new approach to collaborative filtering: Operator estimation with spectral regularization. J. Mach. Learn. Res., 10, 803–826.
2.
AndoR.K., and ZhangT. (2005). A framework for learning predictive structures from multiple tasks and unlabeled data. J. Mach. Learn. Res., 6, 1817–1853.
3.
AshburnerM., BallC.A., BlakeJ.A., et al. (2000). Gene ontology: Tool for the unification of biology. Nat. Gene., 25, 25–29.
4.
BerrizG., KingO., BryantB., et al. (2003). Characterizing gene sets with funcassociate. Bioinformatics, 19, 2502–2504.
5.
CandesE., and RechtB. (2009). Exact matrix completion via convex optimization. Found. Comput. Math., 9, 717.
6.
CaoB., LiuN.N., and YangQ. (2010). Transfer learning for collective link prediction in multiple heterogenous domains. In International Conference on Machine Learning, 159–166.
7.
ChatraryamontriA., CeolA., PelusoD., et al. (2009). Virusmint: A viral protein interaction database. Nucleic Acids Res., 37, D669–D673.
8.
ChenC.-C., LinC.-Y., LoY.-S., and YangJ.-M. (2009). Ppisearch: A web server for searching homologous protein–protein interactions across multiple species. Nucleic Acids Res., 37(suppl 2), W369–W375.
9.
ChenJ., LiuJ., and YeJ. (2012). Learning incoherent sparse and low-rank patterns from multiple tasks. ACM Trans. Knowl. Discov. Data, 5, 22.
10.
ChenJ., TangL., LiuJ., and YeJ. (2013). A convex formulation for learning a shared predictive structure from multiple tasks. IEEE Trans. Pattern Anal. Mach. Intell., 35, 1025–1038.
11.
ChenX., and LiuM. (2005). Prediction of protein-protein interactions using random decision forest framework. Bioinformatics, 21, 4394–4400.
12.
DavisJ., and GoadrichM. (2006). The relationship between precision-recall and roc curves. In Proceedings of the 23rd International Conference on Machine Learning, 233–240.
13.
DyerM., MuraliT., and SobralB. (2007). Computational prediction of host-pathogen protein-protein interactions. Bioinformatics, 23, i159–66.
14.
DyerM., MuraliT.M., SobralB.W. (2011). Supervised learning and prediction of physical interactions between human and hiv proteins. Infect. Genet. Evol., 11, 917–923.
15.
EvgeniouT., and PontilM. (2004). Regularized multi-task learning. In Proceedings of the 10th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 109–117.
JiS., and YeJ. (2009). An accelerated gradient method for trace norm minimization. In Proceedings of the 26th Annual International Conference on Machine Learning, 457–464. ACM.
18.
JinR., SiL., and ZhaiC. (2002). Preference-based graphic models for collaborative filtering. 329–336, In Proceedings of the 19th Conference on Uncertainty in Artificial Intelligence, Morgan Kaufmann Publishers Inc.
19.
KorenY., BellR., and VolinskyC. (2009). Matrix factorization techniques for recommender systems. IEEE Comput., 42, 30–37.
20.
KshirsagarM., CarbonellJ.G., and Klein-SeetharamanJ. (2012). Techniques to cope with missing data in host-pathogen protein interaction prediction. Bioinformatics, 28, i466–i472.
21.
KshirsagarM., CarbonellJ.G., and Klein-SeetharamanJ. (2013). Multi-task learning for host-pathogen protein interactions. Bioinformatics, 29, i217–i226.
22.
KumarR., and NanduriB. (2010). Hpidb—A unified resource for host-pathogen interactions. BMC Bioinformatics, 11(Suppl 6), S16.
23.
LiB., YangQ., and XueX. (2009). Transfer learning for collaborative filtering via a rating-matrix generative model. In International Conference on Machine Learning, 617–624. ACM.
24.
LinC.-Y., ChenY.-C., LoY.-S., and YangJ.-M. (2013). Inferring homologous protein-protein interactions through pair position specific scoring matrix. BMC Bioinformatics, 14, 1.
25.
MatthewsL.R., VaglioP., ReboulJ., GeH., DavisB.P., GarrelsJ., VincentS., and VidalM. (2001). Identification of potential interaction networks using sequence-based searches for conserved protein-protein interactions or interologs. Genome Res., 11, 2120–2126.
26.
MenonA.K., and ElkanC. (2011). Link prediction via matrix factorization. Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 437–452.
27.
MikaS., and RostB. (2006). Protein–protein interactions more conserved within species than across species. PLoS Comput. Biol., 2, e79.
28.
MirzazadehF., WhiteM., GyörgyA., and SchuurmansD. (2015). Scalable metric learning for co-embedding. In Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML-KDD), 625–642. Springer.
29.
MurakamiY. and MizuguchiK. (2014). Homology-based prediction of interactions between proteins using averaged one-dependence estimators. BMC Bioinformatics, 15, 213.
30.
NanboA., ImaiM., WatanabeS., et al. (2010). Ebolavirus is internalized into host cells via macropinocytosis in a viral glycoprotein-dependent manner. PLoS Pathog., 6, e1001121.
31.
NatarajanN., and DhillonI.S. (2014). Inductive matrix completion for predicting gene–disease associations. Bioinformatics, 30, i60–i68.
32.
PhungD.Q., VenkateshS., et al. (2009). Ordinal boltzmann machines for collaborative filtering. In UAI, 548–556. AUAI Press.
33.
QiY., Bar-JosephZ., Klein-SeetharamanJ. (2006). Evaluation of different biological data and computational classification methods for use in protein interaction prediction. Proteins, 63, 490–500.
34.
QiY., DhimanH., Bar-JosephZ., et al. (2009). Systematic prediction of human membrane receptor interactions. Proteomics, 23, 5243–55.
35.
QiY., TastanO., CarbonellJ.G., Klein-SeetharamanJ., and WestonJ. (2010). Semi-supervised multi-task learning for predicting interactions between HIV-1 and human proteins. Bioinformatics, 26, i645–i652.
36.
SarwarB., KarypisG., KonstanJ., and RiedlJ. (2001). Item-based collaborative filtering recommendation algorithms. In Proceedings of the 10th International Conference on World Wide Web, 285–295. ACM.
37.
ShenJ., ZhangJ., LuoX., et al. (2007). Predicting protein-protein interactions based only on sequences information. Proc. Natl. Acad. Sci. U. S. A., 104, 4337–4341.
38.
SinghA.P., and GordonG.J. (2008). Relational learning via collective matrix factorization. In Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 650–658.
39.
SinghR., XuJ., and BergerB. (2006). Struct2net: Integrating structure into protein-protein interaction prediction. Pac. Symp. Biocomput., 11, 403–411.
40.
TastanO., QiY., CarbonellJ.G., et al. (2009). Prediction of interactions between hiv-1 and human proteins by information integration. Pac. Symp. Biocomput.516–527.
41.
TekirS.D., CakirT., and UlgenK.Ö. (2012). Infection strategies of bacterial and viral pathogens through pathogen-human protein protein interactions. Syst. Biol. Microb. Infect., 3, 74.
42.
UniProt Consortium (2011). Ongoing and future developments at the universal protein resource. Nucleic Acids Res., 39, D214–D219.
43.
WalhoutA.J., and VidalM. (2001). Protein interaction maps for model organisms. Nat. Rev. Mol. Cell Biol., 2, 55–63.
44.
WuX., ZhuL., GuoJ., et al. (2006). Prediction of yeast protein-protein interaction network: Insights from the gene ontology and annotations. Nucleic Acids Res., 34, 2137–2150.
45.
XuQ., XiangE.W., and YangQ. (2010). Protein-protein interaction prediction via collective matrix factorization. In International Conference on Bioinformatics and Biomedicine.
ZhouJ., ChenJ., and YeJ. (2011). MALSAR: Multi-task Learning via Structural Regularization. Arizona State University.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
