Snakefile-recalculations

include: "Snakefile"
include: "Snakefile-proteomics"

experiment_type=get_from_config_or_metadata_summary('experiment_type')

# Proteomics experiments might need to be recalculated for new version 
# of bioentities in each Ensembl update
if experiment_type == 'proteomics_baseline' or experiment_type =='proteomics_baseline_dia' or experiment_type =='proteomics_differential':
    print(f"Recalculations not currently implemented for proteomics experiment: {experiment_type}")
    print(f"set goal='reprocess' in the main script to reprocess")
    sys.exit(1)



def get_outputs():
    """
    First method to be executed since it is run by rule all.
    """
    import os.path
    import datetime
    print('Starting getting list of outputs..' + str(datetime.datetime.now()))

    tool_outputs = {}
    tool_outputs['percentile-ranks'] = f"{config['accession']}-percentile-ranks.tsv"
    outputs = []
    #get metrics only for baseline
    global metrics
    if config['tool']=="all-baseline" or 'baseline-tracks' in config['tool'] or 'baseline-heatmap' in config['tool'] or 'baseline-coexpression' in config['tool']:
        metrics = get_metrics_recalculations()

    # Read this now so that it is available for all other needs
    read_metadata_summary()
    global skip_accession
    skip_accession = read_skip_steps_file()
    global run_deconv_accession
    run_deconv_accession = read_run_deconv_file()
    experiment_type=get_from_config_or_metadata_summary('experiment_type')
    required_config=['tool']
    check_config_required(fields=required_config)
    if 'percentile-ranks' in config['tool'] or config['tool']=="all-diff" and skip(config['accession'],'percentile_ranks'):
        outputs.append(f"{config['accession']}-percentile-ranks.tsv")
    if 'differential-tracks' in config['tool'] or config['tool']=="all-diff" and skip(config['accession'],'differential-tracks'):
        check_config_required(fields=['metadata_summary'], method='differential-tracks')
        # fake elements to mix contrasts labels and ids
        outputs.extend(expand(config['accession']+".{id}.{type}", id=get_contrast_ids(), type=["genes.pval.bedGraph", "genes.log2foldchange.bedGraph"]))
    if 'baseline-tracks' in config['tool'] or config['tool']=="all-baseline" and skip(config['accession'],'baseline-tracks'):
        check_config_required(fields=['metadata_summary'], method='baseline-tracks')
        # combine metric (fpkm / tpm) with assay_id/assay_label (zip based)
        # in a product manner
        outputs.extend(expand(config['accession']+".{a_id}.genes.expressions_{metric}.bedGraph",
                            a_id=get_assay_ids(),
                            metric=metrics))
    if 'differential-gsea' in config['tool'] or config['tool']=="all-diff" and skip(config['accession'],'differential-gsea'):
        check_config_required(fields=['bioentities_properties'], method='differential-gsea')
        outputs.extend(
                expand( config['accession']+".{c_id}.{ext_db}.{type}",
                        c_id=get_contrast_ids(),
                        ext_db=get_ext_db(),
                        type=["gsea.tsv", "gsea_list.tsv"]))
        outputs.extend(
                expand("logs/"+config['accession']+".{c_id}.{ext_db}.{type}",
                        c_id=get_contrast_ids(),
                        ext_db=get_ext_db(),
                        type=["check_differential_gsea.done", "check_differential_gsea_list.done"]))
    if 'atlas-experiment-summary' in config['tool'] or 'all' in config['tool'] and skip(config['accession'],'atlas_experiment_summary'):
        outputs.append(f"{config['accession']}-atlasExperimentSummary.Rdata")
    if 'baseline-heatmap' in config['tool'] or 'all-baseline' in config['tool'] and skip(config['accession'],'baseline-heatmap'):
        outputs.extend(expand(f"{config['accession']}"+"-heatmap-{metric}.pdf", metric=metrics ))
        outputs.append(f"{config['accession']}-heatmap.pdf")
    # add deconvolution results for some accessions
    if run_deconvolution(config['accession']):
        outputs.append(f"{config['accession']}-deconvolution.proportions.tsv")
    if 'baseline-coexpression' in config['tool'] or 'all-baseline' in config['tool'] and skip(config['accession'],'baseline-coexpression'):   
        metric_link_coexp=False
        for m in metrics:
            expression_file=f"{config['accession']}-{m}.tsv"
            print(f"Checking file size for {expression_file} and number of columns")
            if os.path.getsize(expression_file) > 0 and get_number_columns(expression_file)>4:
                metric_link_coexp=True
                outputs.extend(expand(f"{config['accession']}-{m}-coexpressions.tsv.gz"))

        if metric_link_coexp == True:
            outputs.extend(expand(f"{config['accession']}-coexpressions.tsv.gz" ))

    print(outputs)
    print('Getting list of outputs.. done')
    print(datetime.datetime.now())

    return outputs


rule all:
    input:
        required_outputs=get_outputs()