84 lines (74 with data), 3.8 kB

################################################
#   Run elastic net regression on each
#   omic-type dataset.
################################################

configfile: 'config.json'

rule:
    input:
        'results/elastic_net_metabolomics.positive_features.tsv',
        'results/elastic_net_metabolomics.negative_features.tsv',
        'results/elastic_net_metabolomics_remove_unknowns.positive_features.tsv',
        'results/elastic_net_metabolomics_remove_unknowns.negative_features.tsv',
        'results/elastic_net_proteomics.positive_features.tsv',
        'results/elastic_net_proteomics.negative_features.tsv',
        'results/elastic_net_lipidomics.positive_features.tsv',
        'results/elastic_net_lipidomics.negative_features.tsv',
        'results/elastic_net_lipidomics_remove_unknowns.positive_features.tsv',
        'results/elastic_net_lipidomics_remove_unknowns.negative_features.tsv',
        'results/regression_analysis/EN_genes.only_COVID.positive_features.tsv',
        'results/regression_analysis/EN_genes.only_COVID.negative_features.tsv'

rule run_transcriptomic_elastic_net:
    input:
        tpm='../raw_data/genes.tpm.no_hg.tab',
        meta='../raw_data/DCD_v3.tsv'
    output:
        'results/regression_analysis/EN_genes.only_COVID.positive_features.tsv',
        'results/regression_analysis/EN_genes.only_COVID.negative_features.tsv'
    run:

        cmds=[
            'mkdir -p {}/regression_analysis'.format(config['output_location']),
            'python elastic_net_hospital_free_transcriptomic.py {{input.tpm}} {{input.meta}} -l -s COVID -o results/regression_analysis/EN_genes.only_COVID'
        ]
        for c in cmds:
            shell('echo "{}"'.format(c))
            shell(c)

rule run_metabolomics_elastic_net:
    output:
        'results/elastic_net_metabolomics.positive_features.tsv',
        'results/elastic_net_metabolomics.negative_features.tsv'    
    run:
        cmd='python regression_hospital_free.py raw_data/2020-06-11_metabolomics.tsv -a elastic_net -o results/elastic_net_metabolomics'
        shell('echo "{}"'.format(cmd))
        shell(cmd)

rule run_metabolomics_elastic_net_remove_unknown:
    output:
        'results/elastic_net_metabolomics_remove_unknowns.positive_features.tsv',
        'results/elastic_net_metabolomics_remove_unknowns.negative_features.tsv'
    run:
        cmd='python regression_hospital_free.py raw_data/2020-06-11_metabolomics.tsv -a elastic_net -r -o results/elastic_net_metabolomics_remove_unknowns'
        shell('echo "{}"'.format(cmd))
        shell(cmd)

rule run_proteomics_elastic_net:
    output:
        'results/elastic_net_proteomics.positive_features.tsv',
        'results/elastic_net_proteomics.negative_features.tsv'
    run:
        cmd='python regression_hospital_free.py raw_data/2020-06-10_proteomics.tsv -a elastic_net -o results/elastic_net_proteomics'
        shell('echo "{}"'.format(cmd))
        shell(cmd)

rule run_lipidomics_elastic_net:
    output:
        'results/elastic_net_lipidomics.positive_features.tsv',
        'results/elastic_net_lipidomics.negative_features.tsv'
    run:
        cmd='python regression_hospital_free.py raw_data/2020-06-10_lipidomics.tsv -a elastic_net -o results/elastic_net_lipidomics'
        shell('echo "{}"'.format(cmd))
        shell(cmd)

rule run_lipidomics_elastic_net_remove_unknown:
    output:
        'results/elastic_net_lipidomics_remove_unknowns.positive_features.tsv',
        'results/elastic_net_lipidomics_remove_unknowns.negative_features.tsv'
    run:
        cmd='python regression_hospital_free.py raw_data/2020-06-10_lipidomics.tsv -a elastic_net -r -o results/elastic_net_lipidomics_remove_unknowns'
        shell('echo "{}"'.format(cmd))
        shell(cmd)