MOTIVATION: Accurate subcategorization of tumour types through gene-expression profiling requires analytical techniques that estimate the number of categories or clusters rigorously and reliably. Parametric mixture modelling provides a natural setting to address this problem. RESULTS: We compare a criterion for model selection that is derived from a variational Bayesian framework with a popular alternative based on the Bayesian information criterion. Using simulated data, we show that the variational Bayesian method is more accurate in finding the true number of clusters in situations that are relevant to current and future microarray studies. We also compare the two criteria using freely available tumour microarray datasets and show that the variational Bayesian method is more sensitive to capturing biologically relevant structure.
MOTIVATION: Accurate subcategorization of tumour types through gene-expression profiling requires analytical techniques that estimate the number of categories or clusters rigorously and reliably. Parametric mixture modelling provides a natural setting to address this problem. RESULTS: We compare a criterion for model selection that is derived from a variational Bayesian framework with a popular alternative based on the Bayesian information criterion. Using simulated data, we show that the variational Bayesian method is more accurate in finding the true number of clusters in situations that are relevant to current and future microarray studies. We also compare the two criteria using freely available tumour microarray datasets and show that the variational Bayesian method is more sensitive to capturing biologically relevant structure.
Authors: Benjamin A Logsdon; James Y Dai; Paul L Auer; Jill M Johnsen; Santhi K Ganesh; Nicholas L Smith; James G Wilson; Russell P Tracy; Leslie A Lange; Shuo Jiao; Stephen S Rich; Guillaume Lettre; Christopher S Carlson; Rebecca D Jackson; Christopher J O'Donnell; Mark M Wurfel; Deborah A Nickerson; Hua Tang; Alexander P Reiner; Charles Kooperberg Journal: Genet Epidemiol Date: 2014-01 Impact factor: 2.135
Authors: Simon Rogers; Mark Girolami; Walter Kolch; Katrina M Waters; Tao Liu; Brian Thrall; H Steven Wiley Journal: Bioinformatics Date: 2008-10-30 Impact factor: 6.937
Authors: Andrew E Teschendorff; Sergio Gomez; Alex Arenas; Dorraya El-Ashry; Marcus Schmidt; Mathias Gehrmann; Carlos Caldas Journal: BMC Cancer Date: 2010-11-04 Impact factor: 4.430
Authors: Yan Jiao; Katherine Lawler; Gargi S Patel; Arnie Purushotham; Annette F Jones; Anita Grigoriadis; Andrew Tutt; Tony Ng; Andrew E Teschendorff Journal: BMC Bioinformatics Date: 2011-10-19 Impact factor: 3.169