Identifying Relevant Group of miRNAs in Cancer using Fuzzy Mutual Information

Authors: Jayanta Kumar Pal, Shubhra Sankar Ray and Sankar K. Pal
jkp_it08@yahoo.com, shubhra@isical.ac.in, sankar@isical.ac.in
Abstract: MicroRNAs (miRNAs) act as a major biomarker of cancer. All miRNAs in human body are not equally important for cancer identification. We propose a novel miRNA selection methodology, called FMIMS, which automatically selects the most relevant miRNAs for a particular type of cancer. In FMIMS, miRNAs are initially grouped by using a SVM based algorithm; then the group with highest relevance is determined and the miRNAs in that group are finally ranked and selected according to their redundancy. Fuzzy mutual information measure is used in computing the relevance of a group and the redundancy of miRNAs within it. Superiority of the most relevant group to all others, in deciding normal or cancer, is demonstrated on breast, renal, colorectal, lung, melanoma and prostate cancer data set. The merit of FMIMS as compared to several existing ones is established. While 12 out of 15 selected miRNAs corroborate with those of biological investigations, three of them are possible novel predictions by FMIMS.
Example Dataset: Lung Cancer (normal.txt, cancer.txt, & miRNA.txt) and Melanoma (normal.txt, cancer.txt, & miRNA.txt)
Supplementary Material
Software
  1. Download the , notmal.txt, cancer.txt and miRNA.txt for any data set and keep all the downloaded files in the same directory.
  2. Run the program.
  3. The program returns the name of the miRNAs. If the name of the miRNAs are not provided (i.e., miRNA.txt file is missing), the program returns the miRNA ID nos. according to the order of the input miRNA expression values.
  4. The program also creats a file "sel.txt" which stores the name or the ID of the selected miRNAs.