ABSTRACT
Non-stationary Dynamic Bayesian Networks (Non-stationary DBNs) are widely used to model the temporal changes of directed dependency structures from multivariate time series data. However, the existing change-points based non-stationary DBNs methods have several drawbacks including excessive computational cost, and low convergence speed. In this paper we proposed a novel non-stationary DBNs method. Our method is based on the perfect simulation model. We applied this approach for network structure inference from synthetic data and biological microarray gene expression data and compared it with other two state-of-the-art non-stationary DBNs methods. The experimental results demonstrated that our method outperformed two other state-of-the-art methods in both computational cost and structure prediction accuracy. The further sensitivity analysis showed that once converged our model is robust to large parameter ranges, which reduces the uncertainty of the model behavior.
Supplemental Material
Available for Download
cikm2012.tex the source file, cikm2012.pdf the pdf file, cikm2012.ps the ps, file fig/ the fold that contains all the fig files
- Chris A. Benedict, Theresa A. Banks, Lionel Senderowicz, Mira Ko, William J. Britt, Ana Angulo, Peter Ghazal, and Carl F. Ware. Lymphotoxins and cytomegalovirus cooperatively induce interferon-b establishing host-virus detente. Immunity, 15:617--626, 2001.Google ScholarCross Ref
- David Maxwell Chickering. Learning equivalence classes of bayesian network structures. Journal of Machine Learning Research, 2:445--498, 2002. Google ScholarDigital Library
- N. Chopin. Dynamic detection of change points in long time series. The Annals of the Institute of Statistical Mathematics, 2006.Google Scholar
- Michael F. Covington, Satchidananda Panda, Xing Liang Liu, Carl A. Strayer, D. Ry Wagner, and Steve A. Kay. Elf3 modulates resetting of the circadian clock in arabidopsis. The Plant Cell, 13:1305--1315, 2001.Google ScholarCross Ref
- Frank Dondelinger, Sophie Lebre, and Dirk Husmeier. Heterogeneous continuous dynamic bayesian networks with flexible structure and inter-time segment information sharing. In Proceedings of 2010 International Conference on Machine Learning (ICML10), 2010.Google Scholar
- Paul Fearnhead. Exact and efficient bayesian inference for multiple changepoint problems. Statistics and Computing, 16:203--213, 2006. Google ScholarDigital Library
- Andrew Gelman and Donald B. Rubin. Inference from iterative simulation using multiple sequences. Statistical Science, 7:457--472, 1992.Google ScholarCross Ref
- Peter J. Green. Reversible jump markov chain monte carlo computation and bayesian model determination. Biometrika, 82:711--732, 1995.Google ScholarCross Ref
- Peter J. Green. Trans-dimensional markov chain monte carlo. Highly Structured Stochastic Systems, Oxford University Press, 2003.Google Scholar
- Marco Grzegorczy, Dirk Husmeier, Kieron D. Edwards, Peter Ghazal, , and Andrew J. Millar. Modelling non-stationary gene regulatory processes with a non-homogeneous bayesian network and the allocation sampler. Bioinformatics, 24:2071 -- 2078, 2008. Google ScholarDigital Library
- Marco Grzegorczyk and Dirk Husmeier. Improvements in the reconstruction of time-varying gene regulatory networks: dynamic programming and regularization by information sharing among genes. Journal of Bioinformatics, 27(5):693--699, 2011. Google ScholarDigital Library
- Anthony Hall, Laszlo Kozma-Bognar, RekaToth, Ferenc Nagy, and Andrew J. Millar. Conditional circadian regulation of phytochrome a gene expression. Plant Physiol.,127(4):1808--18, 2001.Google ScholarCross Ref
- David Heckerman, Dan Geiger, and David Maxwell Chickering. Learning bayesian networks: The combination of knowledge and statistGoogle Scholar
- Kenya Honda, Akinori Takaoka, and Tadatsugu Taniguchi. Type i interferon gene induction by the interferon regulatory factor family of transcription factors. Immunity, 25:349--360, 2006.Google ScholarCross Ref
- Dirk Husmeier. Sensitivity and specificity of inferring genetic regulatory interactions from microarray experiments with dynamic bayesian networks. Bioinformatics, 19:2271--2282, 2003.Google ScholarCross Ref
- Dirk Husmeier, Frank Dondelinger, and Sophie Lebre. Inter-time segment information sharing for non-homogeneous dynamic bayesian networks. In Proceedings of 2010 Advances in Neural Information Processing Systems (NIPS), 2010.Google Scholar
- Katja Ickstadt, Bjorn Bornkamp, Marco Grzegorczyk, Jakob Wieczorek, M. Rahuman Sheriff, Hernan E. Grecco, and Eli Zamir. Nonparametric bayesian networks. In: Bernardo, Bayarri, Berger, Dawid, Heckerman, Smith, and West (eds): Bayesian Statistics 9, Oxford University Press, 2010.Google Scholar
- Yi Jia and Jun Huan. Constructing non-stationary dynamic bayesian networks with a flexible lag choosing mechanism. BMC Bioinformatics, 2010.Google ScholarCross Ref
- JE Darnell Jr, IM Kerr, and GR Stark. Jak-stat pathways and transcriptional activation in response to ifns and other extracellular signaling proteins. Science, 264:1415--1421, 1994.Google ScholarCross Ref
- Yoshinobu Kawahara and Masashi Sugiyama. Change-point detection in time-series data by direct densityratio estimation. In Proceedings of 2009 SIAM International Conference on Data Mining (SDM09), 2009.Google ScholarCross Ref
- M. Lavielle and G. Teyssillree. Adaptive detection of multiple change-points in asset price volatility. Long Memory in Economics, page 129lC156, 2005.Google Scholar
- Sophie Lebre, Jennifer Becq, Frederic Devaux, Michael Stumpf, and Gaelle Lelandais. Statistical inference of the time-varying structure of gene regulation networks. BMC Systems Biology, 4(1):130, 2010.Google ScholarCross Ref
- Robert Lund and Jaxk Reeves. Detection of undocumented changepoints: a revision of the two-phase regression model. J. Climate, 2002.Google Scholar
- Paloma Mas. Circadian clock function in arabidopsis thaliana: time beyond transcription. Trends Cell Biology, 18:273--181, 2008.Google ScholarCross Ref
- Takeshi Mizuno and Norihito Nakamichi. Pseudo-response regulators (prrs) or true oscillator components (tocs). Plant Cell Physiol., 46(5):677--685, 2005.Google ScholarCross Ref
- V. Moskvina and A. Zhigljavsky. An algorithm based on singular spectrum analysis for changepoint detection. Communications in Statistics Part B: Simulation and Computation, 2003.Google Scholar
- Agostino Nobile and Alastair T. Fearnside. Bayesian finite mixtures with an unknown number of components: The allocation sampler. Statistics and Computing, 17:147--162, 2007. Google ScholarDigital Library
- Alessia Para, Eva M. Farre, Takato Imaizumi, Jose L. Pruneda-Paz, Franklin G. Harmon, and Steve A. Kay. Prr3 is a vascular regulator of toc1 stability in the arabidopsis circadian clock. Plant Cell, 19(11):3462--73, 2007.Google ScholarCross Ref
- A. Pievatolo and Peter J. Green. Bounary detection through dynamic polygons. Journal of the Royal Statistical Society, Series B 60:609--626, 1998.Google ScholarCross Ref
- Arvind Rao, Alfred O. Hero III, David J. States, and James Douglas Engel. Inferring time-varying network topologies from gene expression data. EURASIP Journal on Bioinformatics and Systems Biology, 2007. Google ScholarDigital Library
- Sobia Raza, Kevin A Robertson, Paul A Lacaze, David Page, Anton J Enright, Peter Ghazal, and Tom C Freeman. A logic-based diagram of signalling pathways central to macrophage activation. BMC System Biology, 2:Article 36, 2008.Google Scholar
- Joshua W Robinson and Alexander J Hartemink. Non-stationary dynamic bayesian networks. Procedding of Advances in Neural Information Processing Systems Conference, 2008.Google Scholar
- Patrice A. Salome and C. Robertson McClung. The arabidopsis thaliana clock. Journal of Biological Rhythms, 19(5):425--435, 2004.Google ScholarCross Ref
- Le Song, Mladen Kolar, and Eric Xing. Time-varying dynamic bayesian networks. NIPS, 2009.Google Scholar
- Adriano V. Werhli, Marco Grzegorczyk, and Dirk Husmeier. Comparative evaluation of reverse engineering gene ,Bioinformatics, 22 (17):2129--2135, 2006. Google ScholarDigital Library
- Wentao Zhao, Erchin Serpedin, and Edward R. Dougherty. Inferring gene regulatory networks from time series data using the minimum description length principle. Bioinformatics, 22(17):2129-2135, 2006. Google ScholarDigital Library
Index Terms
- Non-stationary bayesian networks based on perfect simulation
Recommendations
Learning the structure of dynamic Bayesian networks from time series and steady state measurements
Dynamic Bayesian networks (DBN) are a class of graphical models that has become a standard tool for modeling various stochastic time-varying phenomena. In many applications, the primary goal is to infer the network structure from measurement data. ...
Simulation-based Bayesian inference for epidemic models
A powerful and flexible method for fitting dynamic models to missing and censored data is to use the Bayesian paradigm via data-augmented Markov chain Monte Carlo (DA-MCMC). This samples from the joint posterior for the parameters and missing data, but ...
Bayesian learning of Bayesian networks with informative priors
This paper presents and evaluates an approach to Bayesian model averaging where the models are Bayesian nets (BNs). A comprehensive study of the literature on structural priors for BNs is conducted. A number of prior distributions are defined using ...
Comments