readme_data.txt

data.mat
	This example data is adapted from Atwell et. al. (2010).
	Genotype and flowering time phenotypes obtained from Atwell et. al. (2010).
        For more info: https://www.ncbi.nlm.nih.gov/pubmed/20336072
	
        -Rare variants (MAF < 0.1) are filtered out.
        -Number of SNPs are reduced to 173219 from 214051.

        X             - {0,1,2} genotype matrix 199 x 173219 (samples vs SNPs)
			To make it compatible with the algorithm we assigned {2} to random locations. 
        Y             - binary phenotype matrix 199 x 17 (samples vs phenotypes) 
			- NaN values are depricated and data is binarized based on the mean.
        genotypes     - genotype matrix 214051 x 2 - 1st column = "0" in X, 2nd column = "1" in X
        MAF           - minor allele frequency vector 173219 x 1 = (# of ones in X) / (# of zeros in X)
        phenotypes    - flowering time phenotypes - 1st column : phenotype_name, 2nd column : index in Atwell (2010) phenotypes
        snp           - SNP information matrix 173219 x 3 - 1st column : snp id, 2nd column: chromosome, 3rd column : position
        samples       - samples information matrix 199 x 2 - Contains 2 different unique index for each sample
	R	      - regulatory/coding information binary vector 173219x1 indicates whether corresponding SNP (in the feature 				matrix) is in regulatory/coding region. (0: not in regulatory/coding region, 1: in regulatory/coding region)
			   
networks.mat
    Contains 4 networks - GS, GM, GI, GS_HICN
                        - GS : Genomic Sequence    - Nearest SNPs are connected
                        - GM : Genomic Membership  - In addition to GS, 
                               SNPs within the same gene are connected as a clique.
                        - GI : Genomic Interaction - In addition to GM,
                               SNPs within interacting genes are connected as a clique.
                               Interacting genes are determined from a Protein-Protein Interaction network.
                        - GS-HICN : In addition to GS, SNPs that are close in 3D are connected.
						
    networks        - 1 x 4 cell array - each cell contains a SNP vs SNP network
                                         i.e. a 173219 x 173219 binary sparse matrix.
								  
    networkNames    - 1 x 4 cell array - contains the names of the networks
	
    network_options - struct - contains various options about network generation step.