Automatic Phenotype Mining Tutorial

Environment
    System :  SUN Solaris 9
     Web Server :  Apache HTTP server 2.0.50
     CGI : Perl 5.8.4
     Programming : MATLAB R14 and MATLAB Webserver 2.0
Input File Format
Plain text numeric format. No missing entry is allowed. No string is allowed. Entries in the same line seperated by space.
  1. The first line: K (number of phenotypes);
  2. The second line: n m (n-number of genes, m-number of samples);
  3. The third line: 1*m vector groundtruth, e.g, 0 0 0 1 1 1. If the groundtruth is unknown, use all "0";
  4. 4 ~ (n+3) lines: numeric data matrix, each line specifies a gene.
Here is a sample data file : 500 genes * 38 samples, 2 phenotypes with groundtruth.

Getting Started
Upload a file:
  1. Click "Browse" button
  2. Select file
  3. Click "Upload Data File" button
Processing Data:
 Click "Processing Data" button

Output Files
  1. Number of Informative Genes:  a number shows how many informative genes was deteted.
  2. Phenotype Partiton of Samples:  a plain text file shows samples' partition.
  3. Informative Genes:  a plain text file shows the indices of the informative genes.
  4. Phenotype Structure Data Matrix:  a plain text file shows the submatrix of the phenotype structure.
  5. Phenotype Structure Image:  a GIF image file shows the phenotype structure.

 

© 2004 Chun Tang. All right reserved. Department of Computer Science and
Engineering, State University of New York at Buffalo