generate.modules: Visualization of chromosome pairwise region epistatic...

In victorcarmelo/WISH: Network construction of whole genome genotype data using the WISH network method

Description

Visualization of chromosome pairwise region epistatic interaction strength, based on statistical significance. The value is based of the most signficant epistatic interaction in each region pair, ranging from 1 ( strongest) to 0 (weakest). By defaulty chromosomes are separated into 1 Mb regions, but if SNPs are more spaced out that this it will adjust to the smallest region that fit the data.

Usage

generate.modules(correlations)

Arguments

correlations	List of epistatic correlations and p-values genrated by epistatic.correlation()
power	Powers to test for creating scale free network. Only change if the default values don't work
n.snps	Number of SNPs to select. SNPs are selected by connectivity, so 500 will select the top 500 most connected Snps. Default is to use all
minClusterSize	Minimum module (cluster) size. Default, is 50, but changing this may be recommended in case of sparse SNPs
type	Type of network to generate. Default is "unsigned", can be "signed" or "signed hybrid"
threads	Number of threads to use if parallelization is possible.

Value

Plots the network connectivity and the scale and SNP tree clustering with modules found. Returns a named list with all the data generated:

• "SNPs"SNPs used in the analysis and their correlations

• "connectivity"The connectivity matrix of the SNPs

• "adjMat"The adjacency matrix of the SNPs

• "dissTom"The dissimilarity TOM

• "genetree"The clustering object used for the genetree

• "modules"The module numbers for each SNP, in order of the SNP matrix

• "modulcolors"The colors used in the modules for each SNP

• "power.estimate"The power estimate to generate a scale free network

Examples

 generate.modules(correlations)

victorcarmelo/WISH documentation built on May 3, 2019, 6:11 p.m.