Table 1

Parameters used for the FCM clustering of the real datasets. Comparison of fuzzification, m, and number of clusters, c, according to the different normalization methods (values in parentheses are the numbers of clusters for the normalized data). N and p indicate the numbers of genes and arrays for the dataset used in the analysis, respectively. In particular, selection of m depends on the normalization method applied to the dataset.




Normalization method




Dataset
N
p
Z (c)
R (c)
Lowess (c)

Serum
517
13
1.25 (10)
1.30 (9)
1.30 (11)
Sporulation
522
16
1.63 (11)
1.75 (11)
1.66 (12)
Yeast
2945
7
1.16 (10)
1.28 (9)
1.66 (13)

Kim et al. BMC Bioinformatics 2006 7:134   doi:10.1186/1471-2105-7-134