The Annals of Applied Statistics
- Ann. Appl. Stat.
- Volume 9, Number 3 (2015), 1643-1670.
The Gibbs-plaid biclustering model
We propose and develop a Bayesian plaid model for biclustering that accounts for the prior dependency between genes (and/or conditions) through a stochastic relational graph. This work is motivated by the need for improved understanding of the molecular mechanisms of human diseases for which effective drugs are lacking, and based on the extensive raw data available through gene expression profiling. We model the prior dependency information from biological knowledge gathered from gene ontologies. Our model, the Gibbs-plaid model, assumes that the relational graph is governed by a Gibbs random field. To estimate the posterior distribution of the bicluster membership labels, we develop a stochastic algorithm that is partly based on the Wang–Landau flat-histogram algorithm. We apply our method to a gene expression database created from the study of retinal detachment, with the aim of confirming known or finding novel subnetworks of proteins associated with this disorder.
Ann. Appl. Stat., Volume 9, Number 3 (2015), 1643-1670.
Received: January 2014
Revised: March 2015
First available in Project Euclid: 2 November 2015
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Chekouo, Thierry; Murua, Alejandro; Raffelsberger, Wolfgang. The Gibbs-plaid biclustering model. Ann. Appl. Stat. 9 (2015), no. 3, 1643--1670. doi:10.1214/15-AOAS854. https://projecteuclid.org/euclid.aoas/1446488755
- Supplement to “The Gibbs-plaid biclustering model”. A high-resolution version of the image shown in Figure 6, as well as the complete biclustering results associated with the RD data have been provided as supplementary material. A proof of the convergence of the stochastic algorithm of Section 3 and further details on Lin’s similarity (Section 2.2) are also included.