Gene Clustering via Integrated Markov Models Combining Individual and Pairwise Features

Abstract

Abstract-Clustering of genes into groups sharing common characteristics is a useful exploratory technique for a number of subsequent computational analysis. A wide range of clustering algorithms have been proposed in particular to analyze gene expression data, but most of them consider genes as independent entities or include relevant information on gene interactions in a sub-optimal way.We propose a probabilistic model that has the advantage to account for individual data (eg. expression) and pairwise data (eg. interaction information coming from biological networks) simultaneously. Our model is based on hidden Markov random field models in which parametric probability distributions account for the distribution of individual data. Data on pairs, possibly reflecting distance or similarity measures between genes, are then included through a graph where the nodes represent the genes and the edges are weighted according to the available interaction information. As a probabilistic model, this model has many interesting theoretical features. Also, preliminary experiments on simulated and real data show promising results and points out the gain in using such an approach.Availability: The software used in this work is written in C++ and is available with other supplementary material at http://mistis.inrialpes.fr/people/forbes/ transparentia/supplementary.html.Index Terms-Markov random fields, model-based clustering, metabolic networks, gene expression.

A Model-Based Approach to Gene Clustering with Missing Observation Reconstruction in a Markov Random Field Framework

“…It was shown to be more efficient than the most widely used clustering approaches on both simulated and real data (Celeux et al, 2003;Vignes and Forbes, 2007). This suggests good properties of convergence; local convergence of a very similar algorithm has been proven in Forbes and Fort (2007).…”

“…This latter method and many others have the drawback to consider gene measurements to be independent. Hence, we proposed in a previous publication (Vignes and Forbes, 2007) an extension of this approach to account for individual features (e.g., microarray data) and dependencies between genes in a united framework based on Hidden Markov Random Fields (HMRF).…”

A fast source-oriented image clustering method for digital forensics

“…Our previous experience of using Markov random field (MRF) approach to image segmentation [53,54] and many others' successful applications of MRF [9,[55][56][57] suggest that the local characteristics of MRFs (also known as Markovianity) allow global optimization problems to be solved iteratively by taking local information into account. Suppose there are K classes of images in the subset, with the value of K unknown, and denote D = {d k |k = 1, 2, .…”

Introduction to Pattern Recognition and Bioinformatics

“…The conventional clustering methods such as hierarchical clustering [139], k-means algorithm [141], self-organizing map [334], principal component analysis [223], graph theoretical approaches [30,36,135,310,372,373,381,384], modelbased clustering [66,103,114,145,222,224,341,352,371,382,383], densitybased approaches [156], fuzzy clustering algorithms [50, 83,113], and rough-fuzzy clustering algorithms [212,213] group coexpressed genes from microarray data. Different supervised gene clustering algorithms are also developed in [84,136,204,205] to find coregulated gene clusters by incorporating the information of sample categories in gene clustering process.…”