Hierarchical Bayesian modelling of gene expression time series across irregularly sampled replicates and clusters

BMC Bioinformatics

Table 3 Results of clustering 300 genes using the four proposed algorithms

	MF	BP	CC	$ℒ$	N. clust.
Agglomerative HGP	0 . 9 2	0 . 3 2	1 . 0	7 3 6 0 . 8	50
agg. GP (as Cooke et al.)	0.78	0.26	0.72	6203.7	128
Mclust (concat.)	0.78	0.14	0.50	1324.0	26
Mclust (averaged)	0.80	0.16	0.42	-736.2	20

The first three columns report the Biological Homogeneity Index (BHI) for the gene ontology categories of Molecular Function (MF), Biological Process (BP) and Cellular component (CC). The two last columns report the log-likelihood $ℒ$ of the models and the number of recovered clusters.

ISSN: 1471-2105