Пример #1
0
def run_stringtie_expression(data):
    """
    estimate expression from Stringtie, using the bcbio datadict
    does not do transcriptome assembly
    """
    bam = dd.get_work_bam(data)
    sample_name = dd.get_sample_name(data)
    out_dir = os.path.join("stringtie", sample_name)
    isoform_fpkm = os.path.join(out_dir, sample_name + ".isoform.fpkm")
    gene_fpkm = os.path.join(out_dir, sample_name + ".fpkm")
    assembly = os.path.abspath(os.path.join(out_dir, "stringtie-assembly.gtf"))
    if file_exists(isoform_fpkm) and file_exists(gene_fpkm):
        data = dd.set_stringtie_dir(data, out_dir)
        data = dd.set_fpkm(data, gene_fpkm)
        data = dd.set_fpkm_isoform(data, isoform_fpkm)
        if "stringtie" in dd.get_transcript_assembler(data):
            assembled_gtfs = dd.get_assembled_gtf(data)
            assembled_gtfs.append(assembly)
            data = dd.set_assembled_gtf(data, assembled_gtfs)
        return data
    with file_transaction(data, out_dir) as tx_out_dir:
        transcript_file = _stringtie_expression(bam, data, tx_out_dir)
        df = _parse_ballgown(transcript_file)
        _write_fpkms(df, tx_out_dir, sample_name)
    data = dd.set_stringtie_dir(data, out_dir)
    data = dd.set_fpkm(data, gene_fpkm)
    data = dd.set_fpkm_isoform(data, isoform_fpkm)
    if "stringtie" in dd.get_transcript_assembler(data):
        assembled_gtfs = dd.get_assembled_gtf(data)
        assembled_gtfs.append(assembly)
        data = dd.set_assembled_gtf(data, assembled_gtfs)
    return data
Пример #2
0
def run_stringtie_expression(data):
    """
    estimate expression from Stringtie, using the bcbio datadict
    does not do transcriptome assembly
    """
    bam = dd.get_work_bam(data)
    sample_name = dd.get_sample_name(data)
    out_dir = os.path.join("stringtie", sample_name)
    isoform_fpkm = os.path.join(out_dir, sample_name + ".isoform.fpkm")
    gene_fpkm = os.path.join(out_dir, sample_name + ".fpkm")
    assembly = os.path.abspath(os.path.join(out_dir, "stringtie-assembly.gtf"))
    if file_exists(isoform_fpkm) and file_exists(gene_fpkm):
        data = dd.set_stringtie_dir(data, out_dir)
        data = dd.set_fpkm(data, gene_fpkm)
        data = dd.set_fpkm_isoform(data, isoform_fpkm)
        if "stringtie" in dd.get_transcript_assembler(data):
            assembled_gtfs = dd.get_assembled_gtf(data)
            assembled_gtfs.append(assembly)
            data = dd.set_assembled_gtf(data, assembled_gtfs)
        return data
    with file_transaction(data, out_dir) as tx_out_dir:
        transcript_file = _stringtie_expression(bam, data, tx_out_dir)
        df = _parse_ballgown(transcript_file)
        _write_fpkms(df, tx_out_dir, sample_name)
    data = dd.set_stringtie_dir(data, out_dir)
    data = dd.set_fpkm(data, gene_fpkm)
    data = dd.set_fpkm_isoform(data, isoform_fpkm)
    if "stringtie" in dd.get_transcript_assembler(data):
        assembled_gtfs = dd.get_assembled_gtf(data)
        assembled_gtfs.append(assembly)
        data = dd.set_assembled_gtf(data, assembled_gtfs)
    return data