preprocess_raw_biom_matrix.conseq

rule process_expr_matrix:
    inputs: in={"type": "raw-expr-matrix"}, 
        script=fileref("scripts/process_expr_matrix.py"),
        hdf5_utils=fileref('scripts/hdf5_utils.py', copy_to='hdf5_utils.py'),
        cleanup_script=fileref("scripts/cleanup_hdf5.py"),
        meta={'type': 'cell_line_metadata'}
    outputs: {
      'type': 'biomarker-matrix',
      'source_dataset_id' : '{{ inputs.in.dataset_id }}',
      'category': '{{inputs.in.category}}',
      'filename': {"$filename": "out-c.hdf5"}
    }
    run "python3 {{inputs.script.filename}} {{ config.SCRIPT_DIR }} {{ inputs.in.dataset_id }} out.hdf5"
    run "python3 {{inputs.cleanup_script.filename}} out.hdf5 out-c.hdf5 --add-arxspan-to-cols -m {{inputs.meta.filename}}"

rule download_biomarker_as_hdf5:
    inputs: 
        in={"type": "biomarker-correctly-transposed"},
        meta={'type': 'cell_line_metadata'},
        hdf5_utils=fileref('scripts/hdf5_utils.py', copy_to='hdf5_utils.py'),
        cleanup_script=fileref("scripts/cleanup_hdf5.py")
    outputs: {
      "type": "biomarker-matrix",
      "source_dataset_id" : "{{ inputs.in.dataset_id }}",
      "category": "{{ inputs.in.category }}",
      "filename": {"$filename": "{{ inputs.in.category }}-c.hdf5"}
    }
    run "python" with """
        import hdf5_utils

        from taigapy import create_taiga_client_v3

        tc = create_taiga_client_v3()

        x = tc.get("{{ inputs.in.dataset_id }}")
        hdf5_utils.write_hdf5(x, "{{inputs.in.category}}.hdf5")
    """
    run "python3 {{inputs.cleanup_script.filename}} {{inputs.in.category}}.hdf5 {{inputs.in.category}}-c.hdf5 --add-arxspan-to-cols -m {{inputs.meta.filename}}"

rule transpose_biomarker:
    inputs: 
        in={"type": "biomarker-needing-transpose"},
        cleanup_script=fileref("scripts/cleanup_hdf5.py"),
        hdf5_utils=fileref('scripts/hdf5_utils.py', copy_to='hdf5_utils.py'),
        meta={'type': 'cell_line_metadata'}
    outputs: {
      "type": "biomarker-matrix",
      "source_dataset_id" : "{{ inputs.in.dataset_id }}",
      "category": "{{ inputs.in.category }}",
      "filename": {"$filename": "{{inputs.in.category}}-c.hdf5"}
    }
    run "python3" with """
        from hdf5_utils import write_hdf5
        from taigapy import create_taiga_client_v3
        import pandas as pd

        tc = create_taiga_client_v3()
        data = tc.get("{{inputs.in.dataset_id}}")
        transposed_data = data.transpose()
        write_hdf5(transposed_data, "{{inputs.in.category}}.hdf5")
    """
    # it appears that the --add-arxspan-to-cols works fine even if the cols are already arxspan
    run "python3 {{inputs.cleanup_script.filename}} {{inputs.in.category}}.hdf5 {{inputs.in.category}}-c.hdf5 --add-arxspan-to-cols -m {{inputs.meta.filename}}"

rule process_maf_file:
    inputs: in={"type": "mutation-maf"}, meta={'type': 'cell_line_metadata'}
    outputs: {"type": "mutation-maf-file", "filename": {"$filename": "maf.csv"}, "orig_dataset_id": "{{inputs.in.dataset_id}}"}
    run "python3" with """
        from taigapy import create_taiga_client_v3
        import pandas as pd

        tc = create_taiga_client_v3()
        data = tc.get("{{inputs.in.dataset_id}}")
        data.to_csv("maf.csv", index=False)
    """

rule process_mutations_matrices:
    inputs:
        mutation_bool={"type": "raw-mutations-bool-matrix"},
        hgnc={"type": "hgnc-snapshot"},
        script=fileref("scripts/download_mutation_boolean_matrix.py"),
        hdf5_script=fileref("scripts/hdf5_utils.py")
    outputs:
        {
            "type": "biomarker-matrix",
            "source_dataset_id": "{{ inputs.mutation_bool.dataset_id }}",
            "category": "mutations-{{ inputs.mutation_bool.category }}",
            "filename": { "$filename": "out.hdf5" }
        }
    run """python {{ inputs.script.filename }} {{ inputs.mutation_bool.dataset_id }} \
                  {{ inputs.hgnc.dataset_id }} out.csv"""
    run "python {{ inputs.hdf5_script.filename }} to_hdf5 out.csv csv out.hdf5"

rule create_fusions_matrix:
    inputs: fusions={'type': 'other-taiga-dataset', 
                'category': 'fusions'},
            script=fileref("scripts/make_fusions_matrix.py"),
            hgnc={"type": "hgnc-snapshot"},
            hdf5_script=fileref("scripts/hdf5_utils.py")
    outputs: {'type': 'biomarker-matrix', 
                'category': 'fusions', 
                'filename': {"$filename": "fusions.hdf5"}, 
                'source_dataset_id': '{{inputs.fusions.dataset_id}}'}
    run """python {{ inputs.script.filename }} {{ inputs.fusions.dataset_id }} \
                  {{ inputs.hgnc.dataset_id }} out.csv"""
    run "python {{ inputs.hdf5_script.filename }} to_hdf5 out.csv csv fusions.hdf5"