Split Guppy sequencing summaries by barcode

split_summary_by_barcode.py

def split_summary_by_barcode(summary_path, out_dir, run_name):
    '''Given a sequencing summary file path, write per barcode summaries to an output directory'''
    
    dtypes = {
        'filename_fastq': 'object',
        'filename_fast5': 'object',
        'read_id': 'object',
        'run_id': 'category',
        'channel': 'int64',
        'mux': 'int64',
        'start_time': 'float64',
        'duration': 'float64',
        'num_events': 'int64',
        'passes_filtering': 'bool',
        'template_start': 'float64',
        'num_events_template': 'int64',
        'template_duration': 'float64',
        'sequence_length_template': 'int64',
        'mean_qscore_template': 'float64',
        'strand_score_template': 'float64',
        'median_template': 'float64',
        'mad_template': 'float64',
        'pore_type': 'category',
        'experiment_id': 'category',
        'sample_id': 'category',
        'end_reason': 'category',
        'alias': 'category',
        'type': 'category',
        'barcode_arrangement': 'category',
        'barcode_full_arrangement': 'category',
        'barcode_kit': 'category',
        'barcode_variant': 'category',
        'barcode_score': 'float64',
        'barcode_front_id': 'category',
        'barcode_front_score': 'float64',
        'barcode_front_refseq': 'category',
        'barcode_front_foundseq': 'category',
        'barcode_front_foundseq_length': 'int64',
        'barcode_front_begin_index': 'int64',
        'barcode_rear_id': 'category',
        'barcode_rear_score': 'float64',
        'barcode_rear_refseq': 'object',
        'barcode_rear_foundseq': 'object',
        'barcode_rear_foundseq_length': 'int64',
        'barcode_rear_end_index': 'int64',
        'bc_front': 'category',
        'bc_rear': 'category'}
    
    os.makedirs(out_dir, exist_ok=True)

    df = pd.read_csv(summary_path, sep='\t', dtype=dtypes)
    gb = df.groupby('barcode_arrangement')
    for barcode in gb.groups:
        gb.get_group(barcode).to_csv(f'{out_dir}/{run_name}_{barcode}.txt', sep='\t', index=False)