Table 4 Parameter settings of biclustering algorithms applied to the Spellman dataset [20].
From: Construction of gene regulatory networks using biclustering and bayesian networks
Algorithm | Parameters | Parameter Description |
|---|---|---|
ISA | tg = 2.0 | Gene threshold level |
| Â | tc = 2.0 | Condition threshold level |
| Â | SN = 500 | No of seeds |
CC | Delta = 0.5 | Maximum accepted score |
| Â | Alpha = 1.2 | Scaling factor |
| Â | M = 100 | Number of biclusters to be found |
OPSM | l = 100 | Number of passed models for each iteration |
K-means | M = 100 | Number of biclusters to be found |
| Â | IN = 100 | Number of Iterations |
| Â | RN = 10 | Number of replications |
| Â | DM = ED | Distance Metric is Euclidean Distance |
Bivisu | NT = 0.5819 | Data noise threshold |
| Â | % NR = 1.57 | Minimum% of rows |
| Â | NC = 5 | Minimum number of columns |
| Â | O% = 25% | Maximum overlap allowed |
MSBE | alpha = 0.4 | Similarity threshold |
| Â | beta = 0.5 | Bonus similarity threshold |
| Â | gamma = 1.2 | Threshold of the average similarity score |
SAMBA | MHS = 100 | Maximal memory allocated for hashing |
| Â | KHS1 = 4 | stage |
| Â | PC = 100 | Maximal kernel size in the hashing stage |
| Â | KHS2 = 4 | Minimal number of responding probes per condition |
| Â | O% = 25% | Minimal kernel size in the hashing stage Maximum overlap between two biclusters |