Construction of gene regulatory networks using biclustering and bayesian networks

Table 2 Statistical comparison of bicluster outputs when using the Gasch Dataset [19].

Biclustering Algorithm	No of Biclusters	Bicluster/Cluster Size		GeneCoverage%	ConditionCoverage %
		Min	Max
ISA	9	50 × 35	155 × 37	25	97
CC	69	11 × 5	2259 × 134	100	100
OPSM	2	11 × 15	575 × 6	88.5	32.9
BiVisu	100	27 × 142	99 × 52	55	100
Kmeans	100	20 × 173	50 × 173	100	100

These biclusters were produced when biclustering algorithms were applied with the parameter settings in Table 1 on the Gasch et al. [19] dataset. The size of a bicluster is determined by its number of values, i.e. the product of the numbers of rows and columns; GeneCoverage% and ConditionCoverage% are the percentage of genes, and conditions were recovered for each algorithm respectively.

ISSN: 1742-4682