DGEclust is a program for clustering and differential expression analysis of digital expression data generated by next-generation sequencing assays, such as RNA-seq, CAGE and others. It takes as input a table of count data and it estimates the number and parameters of the clusters supported by the data. At a later stage, these can be used for identifying differentially expressed genes and for gene- and sample-wise clustering of the original data matrix. Internally, DGEclust uses a Hierarchical Dirichlet Process Mixture Model for modelling over-dispersed count data, combined with a blocked Gibbs sampler for efficient Bayesian learning.
This program is part of the software collection of the Computational Genomics Group at the University of Bristol and it is under continuous development. You can find more technical details on the statistical methodologies used in this software in the following papers:
- http://arxiv.org/abs/1301.4144 (Vavoulis & Gough, J Comput Sci Syst Biol 7:001-009, 2013)
- http://arxiv.org/abs/1405.0723 (Vavoulis et al., submitted, 2014)
For more information, including bug reports, send an email to Dimitris.Vavoulis@bristol.ac.uk or Julian.Gough@bristol.ac.uk
Enjoy!