fxchrom package includes the Matlab program to perform association test for X-chromosomal genetic variants.
The package is currently available to be downloaded here.
Syntax
Output = fxchrom(N,datafile)
Output = fxchrom(N,datafile,covfile)
Description
Program includes three Matlab scripts. 'fxchrom.m' is the function to execute the program. 'fLLR.m' is the function to calculate the log-likelihood ratio which corresponds to the best fitted X-chromosome inactivation model, which is called by 'fxchrom.m' and 'fperm.m'. 'fperm.m' is the function to calculate the permuted log-likelihood ratios for evaluation of the empirical p value, which is called by 'fxchrom.m'.
Output = fxchrom(N,datafile) returns a structure of association test results based on N permutations using a data set read from datafile.
Output = fxchrom(N,datafile,covfile) additionally allows you to input another file for covariates.
The 'datafile' is a .txt file (tab-delimited ASCII files) with four columns: case-control status, sex and genotypes (2 columns). The 'covfile' is a .txt file (tab-delimited ASCII files).
fxchrom treats NaN in the 'datafile' and 'covfile' as missing values and ignores them.
Output is a structure with the following fields:
model — The best fitted X-chromosome inactivation model. Currently, it returns four different models: skewed XCI to normal allele, skewed XCI to risk allele, random XCI and escape of XCI.
beta — Coefficient estimate corresponding to the best fitted XCI model
se — Standard error of the coefficient estimate corresponding to the best fitted XCI model
p — Empirical p value based on N permutations
Examples:
Sample files:
File 'data.txt': sample data file.
File 'covariates.txt': sample covariates file (2 covariates)
Sample codes:
N=100;
datafile='data.txt';
covfile='covariates.txt';
fxchrom(N,datafile)
fxchrom(N,datafile,covfile)
Sample outcomes:
model: 'skewed XCI to risk allele'
beta: 0.0589
se: 0.0473
LLR: 1.5508
p: 0.3700
Reference:
Wang J, Shete S. X-chromosome genetic association test accounting for X-inactivation, skewed X-inactivation, and escape from X-inactivation. Genet Epidemiol. Submitted.