ClusterPSO
Software instruction ClusterPSO version 1 was presented in March 15th, 2014, this program is a free software.
Please report the bugs or suggest for any improvements of current software to the authors. (Author: Yi-Cheng Chiang, E-mail: a09210917@yahoo.com.tw)
This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Input parameters To run the program, input the command "java -jar CpGI_fat.jar contig_sequence data_path threshold_dt p-value population generation"
- contig_sequence is the contig sequence which is defined as 0 or 1. If user want to predict CpG islands from contig sequence that user will set 0, otherwise set 1.
- data_path is the data set file pathway and name (only accept .txt extension, and the input does not have the file extension). For the supplementary test data set, use "D:\CpG_jar\NT_028395.3.fasta".
- threshold_dt is the distance threshold between adjacent CpG dinucleotides within array.
- p-value is the significance level (e.g. 0.01) for the cluster, which indicates a potential for the presence of island within the detected cluster.
- population is the population size of particle swarm optimization.
- generation is the generation number of particle swarm optimization.
Example for ClusterPSO v1 implementation: java -jar CpGI_fat.jar 0 D:\CpG_jar\NT_028395.3.fasta 65 0.01 300 100 The example shows the implementation of the ClusterPSO on contig sequence of 0 using supplementary test data set placed in "D:\CpG_jar\NT_028395.3.fasta", with the distance threshold set as 65%, significance level at 0.01, total test by 300 population and 100 generation in particle swarm optimization.