Skip to main content

Table 1 Parameter settings for biclustering algorithms and post-filtering in the experiments on artificial datasets

From: Identification of coherent patterns in gene expression data using an efficient biclustering algorithm and parallel coordinate visualization

Experiment Algorithm/post-filtering Parameter settings*
Artificial datasets for additive models PA ε = 0.5 – 2.0, N r = 21, N c = 5, P o = 50
  C&C δ = 0.04 – 0.5, α = 1.2, M = 40
  pCluster δ = 0.5 – 1.0, N r = 21, N c = 5
  Post-filtering N r = 21, N c = 5, P o = 50 and M = 10
Artificial datasets for multiplicative models PM ε = 0.2 – 0.6, N r = 18, N c = 4, P o = 25
  PAL ε = 0.4 – 1.0, N r = 18, N c = 4, P o = 25
  C&C δ = 0.04 – 0.5, α = 1.2, M = 20
  pCluster δ = 0.5 – 1.0, N r = 18, N c = 4
  Post-filtering N r = 18, N c = 4, P o = 25 and M = 5
  1. * The definitions of parameters ε, N r , N c and P o follow those defined for the proposed algorithm, i.e. noise threshold, minimum number of rows, minimum number of columns and maximum percentage in overlap allowed in biclusters respectively. Furthermore, M denotes the maximum number of biclusters required and δ of C&C and the pCluster algorithm is defined as in the original publications [12, 25].