paths:
    output_dir: "/doctorai/skanduri/fdr_hacking_pilot/v1_analyses"
    realworld_data: "/doctorai/skanduri/fdr_hacking_pilot/fdr_hacking/data/realworld_methyl_beta.h5"
simulating:
    with_dependencies_beta_features_10k:
        n_observations: [20, 100, 200] # number of examples
        n_sites: [10000]
        dependencies: [1]
        correlation_strength: ["high", "medium"] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [0.3, 0.2] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [[0, 1, 10, 100, 1000, 10000]]
        n_datasets: 10000
        data_distribution: ["beta"]
    without_dependencies_beta_features_10k:
        n_observations: [20, 100, 200] # number of examples
        n_sites: [10000]
        dependencies: [0]
        correlation_strength: ["not_applicable"] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [0.3] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [[0, 1, 10, 100, 1000, 10000]]
        n_datasets: 10000
        data_distribution: ["beta"]
    with_dependencies_beta_features_1k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 1000 ]
        dependencies: [ 1 ]
        correlation_strength: [ "high", "medium" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3, 0.2] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000] ]
        n_datasets: 10000
        data_distribution: [ "beta" ]
    without_dependencies_beta_features_1k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 1000 ]
        dependencies: [ 0 ]
        correlation_strength: [ "not_applicable" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3 ] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000] ]
        n_datasets: 10000
        data_distribution: [ "beta" ]
    with_dependencies_normal_features_10k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 10000 ]
        dependencies: [ 1 ]
        correlation_strength: [ "high", "medium" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3, 0.2] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000, 10000 ] ]
        n_datasets: 10000
        data_distribution: [ "normal" ]
    without_dependencies_normal_features_10k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 10000 ]
        dependencies: [ 0 ]
        correlation_strength: [ "not_applicable" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3 ] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000, 10000 ] ]
        n_datasets: 10000
        data_distribution: [ "normal" ]
    with_dependencies_normal_features_1k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 1000 ]
        dependencies: [ 1 ]
        correlation_strength: [ "high", "medium" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3, 0.2] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000] ]
        n_datasets: 10000
        data_distribution: [ "normal" ]
    without_dependencies_normal_features_1k:
        n_observations: [ 20, 100, 200] # number of examples
        n_sites: [ 1000 ]
        dependencies: [ 0 ]
        correlation_strength: [ "not_applicable" ] # the correlation coefficients in one of the triangles in a correlation matrix will be sampled equally from the ranges supplied here
        bin_size_ratio: [ 0.3 ] # remember to have just one value for this and n_sites
        reporting_histogram_bins: [ [ 0, 1, 10, 100, 1000] ]
        n_datasets: 10000
        data_distribution: [ "normal" ]
statistical_testing:
    beta:
        statistical_test: [ 't-test', 'limma', 'rank-sum' ]
        multipletest_correction_type: ['bonferroni', 'bh' ]
        alpha: [0.01, 0.05, 0.1]
    normal:
        statistical_test: ['ks-test']
        multipletest_correction_type: [ 'bonferroni', 'bh' ]
        alpha: [ 0.01, 0.05, 0.1 ]
reporting:
    compare_sample_size_beta:
        n_sites: [10000]
        correlation_strength: ["high", "not_applicable"]
        bin_size_ratio: [0.3]
        statistical_test: ["t-test"]
        multipletest_correction_type: ["bh"]
        alpha: [0.05]
        data_distribution: ["beta"]
        x_axis: "n_observations"
    compare_sample_size_normal:
        n_sites: [ 10000 ]
        correlation_strength: [ "high", "not_applicable" ]
        bin_size_ratio: [ 0.3 ]
        statistical_test: [ "ks-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "normal" ]
        x_axis: "n_observations"
    compare_feature_size_beta:
        n_observations: [100]
        correlation_strength: [ "high", "not_applicable" ]
        bin_size_ratio: [ 0.3 ]
        statistical_test: [ "t-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "beta" ]
        x_axis: "n_sites"
    compare_feature_size_normal:
        n_observations: [100]
        correlation_strength: [ "high", "not_applicable" ]
        bin_size_ratio: [ 0.3 ]
        statistical_test: [ "ks-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "normal" ]
        x_axis: "n_sites"
    compare_correlation_strength_beta:
        n_observations: [100]
        n_sites: [ 10000 ]
        dependencies: [ 1 ]
        bin_size_ratio: [ 0.3 ]
        statistical_test: [ "t-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "beta" ]
        x_axis: "correlation_strength"
    compare_correlation_strength_normal:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        dependencies: [ 1 ]
        bin_size_ratio: [ 0.3 ]
        statistical_test: [ "ks-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "normal" ]
        x_axis: "correlation_strength"
    compare_binsize_ratio_beta:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        dependencies: [ 1 ]
        correlation_strength: [ "high"]
        statistical_test: [ "t-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "beta" ]
        x_axis: "bin_size_ratio"
    compare_binsize_ratio_normal:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        dependencies: [ 1 ]
        correlation_strength: [ "high" ]
        statistical_test: [ "ks-test" ]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        data_distribution: [ "normal" ]
        x_axis: "bin_size_ratio"
    compare_statistical_test:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        bin_size_ratio: [0.3]
        correlation_strength: ["high", "not_applicable"]
        multipletest_correction_type: [ "bh" ]
        alpha: [ 0.05 ]
        x_axis: "statistical_test"
    compare_multipletest_correction_type_beta:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        bin_size_ratio: [ 0.3 ]
        correlation_strength: [ "high", "not_applicable" ]
        statistical_test: [ "t-test" ]
        alpha: [ 0.05 ]
        data_distribution: [ "beta" ]
        x_axis: "multipletest_correction_type"
    compare_multipletest_correction_type_normal:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        bin_size_ratio: [ 0.3 ]
        correlation_strength: [ "high", "not_applicable" ]
        statistical_test: [ "ks-test" ]
        alpha: [ 0.05 ]
        data_distribution: [ "normal" ]
        x_axis: "multipletest_correction_type"
    compare_alpha_beta:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        bin_size_ratio: [ 0.3 ]
        correlation_strength: [ "high", "not_applicable" ]
        statistical_test: [ "t-test" ]
        multipletest_correction_type: [ "bh" ]
        data_distribution: [ "beta" ]
        x_axis: "alpha"
    compare_alpha_normal:
        n_observations: [ 100 ]
        n_sites: [ 10000 ]
        bin_size_ratio: [ 0.3 ]
        correlation_strength: [ "high", "not_applicable" ]
        statistical_test: [ "ks-test" ]
        multipletest_correction_type: [ "bh" ]
        data_distribution: [ "normal" ]
        x_axis: "alpha"