CGA-MWS-algorithm

JAVA source code of CGA-MWS algorithm

A brief description of algorithm

• INPUT: a weighted non-binary mutation matrix A, a parameter K;
• OUTPUT: a submatrix M.

Example of txt file input to algorithm

Gene	TP53	CDKN2A	CDKN2B	RB1	CDK4	…
Sample_1	1	1.5	0	0	1	…
Sample_2	0.45	1	0	0.4	1.5	…
Sample_3	0	1.5	0	0.3	0	…
…	…	…	…	…	…	…

The process of executing the project

1. You need to import the downloaded my folder, sample files GBM_ GeneNumbers_ 920.txt and GBM_ removeGene_ GeneNumbers_ 911.txt into eclipse or MyEclipse and execute them in the JAVA8 environment whenever it is possible. Files are stored as follows:
   

   

2. The main method in Run.java is the entry to the whole program.

3. Enter the relative or absolute path of the txt file in the following statement.

   String path = "GBM_removeGene_GeneNumbers_911.txt;";
   

4. Setting parameters.

   • This project provides two real data of GBM, GBM_ GeneNumbers_ 920.txt is a file processed according to the paper description, which includes 90 samples and 920 genes. And GBM_ removeGene_ GeneNumbers_ 911.txt is a file that deleted the genes mentioned in the paper when K=4 is tested. The parameter g is given in the file name.

   • If the input sample file GBM_removeGene_GeneNumbers_911.txt is used, the parameter K can to be modified in [4,6] and other parameters remain the default in the r.run() statement.

   • If the input sample file GBM_removeGene_GeneNumbers_920.txt is used, the parameter K can to be modified in [2,3], the parameter g needs to be changed to 920 and other parameters remain the default in the r.run() statement. You should modify the parameters as follows:

     int g = 911;
     int K = 6;
     r.run(paths, g, K, g / 2, 1000, 0.3, 1, 1000, "calfitness_Cov");
     

     • The first parameter: The path of txt file，
     • The second parameter: Number of genes in txt file，
     • The third parameter: The size of Gene set (K)，
     • The fourth parameter: Population size (N, this value is the size of all populations combined)，
     • The fifth parameter: Iteration steps (maxg)，
     • The sixth parameter: Mutation probability (Pm)，
     • The seventh parameter: Number of times the algorithm is executed，
     • The eighth parameter: Number of cycles when calculating p-value,
     • The ninth parameter: Model name ("calfitness_Cov"),

5. After setting the parameters, CGA-MWS algorithm is ready to be executed.

Some supplementary notes

• If you input other custom file, please check the file format and adjust the parameters.

• When the input sample file GBM_removeGene_GeneNumbers_911.txt is used, and all parameters are by default, the printout of CGA-MWS algorithm after successful execution is as follows:

  NO.1  Execute the algorithm
  total time：0.5211s
  
  The best solution is obtained at the No.1 execution time, and the gene set is:{ PTEN, EGFR, PIK3R1, COL1A2, PDGFRA, PIK3CA, }
  Fitness: 209.6769
  CO(M): 70.5
  ME(M): 139.1769
  
  The average running time of (1) times executions is: 0.5211s
  p-value is: 1.0
  

• The seventh parameter determines how many times to execute the algorithm. If you need to set different values, here is the output print when the parameter is 3:

  NO.1  Execute the algorithm
  total time：0.598s
  
  NO.2  Execute the algorithm
  total time：0.422s
  
  NO.3  Execute the algorithm
  total time：0.248s
  
  The best solution is obtained at the No.1 execution time, and the gene set is:{ PTEN, EGFR, PIK3R1, PIK3CA, PDGFRA, COL1A2, }
  Fitness: 209.6769
  CO(M): 70.5
  ME(M): 139.1769
  
  The average running time of (3) times executions is: 0.4227s
  p-value is: 1.0