Burden shift analysis
From burden test results, CWAS-Plus can find overrepresented domains across significant categories. By comparing the observed number of significant categories with the number of significant categories from samples with a randomly shuffled phenotype, the significance of the overrepresentation is determined.
The parameters of the command are as below:
-i, –input_file: Path to the input file which is the result of binomial burden test (*.burden_test.txt).
-b, –burden_res: Path to the result of burden shift from permutation test (*.binom_pvals.txt.gz).
-c_info, –category_info: Path to a text file with category information (*.category_info.txt).
-o_dir, –output_directory: Path to the directory where the output files will be saved. By default, outputs will be saved at
$CWAS_WORKSPACE
.-c_set, –cat_set: Path to the category information file from binomial burden test (*.category_info.txt).
-c_count, –cat_count: Path of the categories counts file from binomial burden test (*.category_counts.txt).
-t, –tag: Tag used for the name of the output files. By default, None.
-c_cutoff, –count_cutoff: The number of cutoff for category counts. It must be positive value. By default, 7.
–pval: P-value threshold. By default, 0.05.
cwas burden_shift -i INPUT.burden_test.txt \
-b INPUT.binom_pvals.txt.gz \
-o_dir OUTPUT_DIR \
-c_info INPUT.category_info.txt \
-c_count INPUT.category_counts.txt \
-c_cutoff 7 \
--pval 0.05