def alignment_workflow(args):
config = inpututils.load_config(args)
config = config['alignment']
lib = args["library_id"]
alignment_dir = args["out_dir"]
bams_dir = args["bams_dir"]
trim = args['trim']
center = args['sequencing_center']
sampleinfo = inpututils.get_sample_info(args['input_yaml'])
cellids = inpututils.get_samples(args['input_yaml'])
fastq1_files, fastq2_files = inpututils.get_fastqs(args['input_yaml'])
alignment_files = get_output_files(alignment_dir, lib)
alignment_meta = os.path.join(alignment_dir, 'metadata.yaml')
bam_files_template = os.path.join(bams_dir, '{cell_id}.bam')
mt_bam_files_template = os.path.join(bams_dir, '{cell_id}_MT.bam')
bams_meta = os.path.join(bams_dir, 'metadata.yaml')
lanes = sorted(set([v[1] for v in fastq1_files.keys()]))
cells = sorted(set([v[0] for v in fastq1_files.keys()]))
input_yaml_blob = os.path.join(alignment_dir, 'input.yaml')
workflow = pypeliner.workflow.Workflow()
workflow.setobj(
obj=mgd.OutputChunks('cell_id', 'lane'),
value=list(fastq1_files.keys()),
)
workflow.subworkflow(
name='alignment_workflow',
func=align.create_alignment_workflow,
args=(
mgd.InputFile('fastq_1',
'cell_id',
'lane',
fnames=fastq1_files,
axes_origin=[]),
mgd.InputFile('fastq_2',
'cell_id',
'lane',
fnames=fastq2_files,
axes_origin=[]),
mgd.OutputFile('bam_markdups',
'cell_id',
template=bam_files_template,
axes_origin=[],
extensions=['.bai']),
mgd.OutputFile('mt_bam_markdups',
'cell_id',
template=mt_bam_files_template,
axes_origin=[],
extensions=['.bai']),
mgd.OutputFile(alignment_files['alignment_metrics_csv']),
mgd.OutputFile(alignment_files['gc_metrics_csv']),
mgd.OutputFile(alignment_files['fastqc_metrics_csv']),
mgd.OutputFile(alignment_files['plot_metrics_output']),
config['ref_genome'],
config,
sampleinfo,
cellids,
mgd.OutputFile(alignment_files['alignment_metrics_tar']),
lib,
trim,
center,
),
)
workflow.transform(
name='generate_meta_files_results',
func='single_cell.utils.helpers.generate_and_upload_metadata',
args=(sys.argv[0:], alignment_dir, list(alignment_files.values()),
mgd.OutputFile(alignment_meta)),
kwargs={
'input_yaml_data': inpututils.load_yaml(args['input_yaml']),
'input_yaml': mgd.OutputFile(input_yaml_blob),
'metadata': {
'library_id': lib,
'cell_ids': cells,
'lane_ids': lanes,
'type': 'alignment'
}
})
workflow.transform(
name='generate_meta_files_bams',
func='single_cell.utils.helpers.generate_and_upload_metadata',
args=(sys.argv[0:], bams_dir,
mgd.Template('aligned.bam',
'cell_id',
template=bam_files_template),
mgd.OutputFile(bams_meta)),
kwargs={
'metadata': {
'library_id': lib,
'cell_ids': cells,
'lane_ids': lanes,
'type': 'cellbams'
},
'template':
(mgd.InputChunks('cell_id'), bam_files_template, 'cell_id'),
})
return workflow
def hmmcopy_workflow(args):
config = inpututils.load_config(args)
config = config['hmmcopy']
sampleinfo = inpututils.get_sample_info(args['input_yaml'])
cellids = inpututils.get_samples(args['input_yaml'])
bam_files = inpututils.get_bams(args['input_yaml'])
lib = args["library_id"]
workflow = pypeliner.workflow.Workflow(
ctx={'docker_image': config['docker']['single_cell_pipeline']}, )
hmmcopy_dir = args["out_dir"]
hmmcopy_files = get_output_files(hmmcopy_dir, lib)
hmmcopy_meta = os.path.join(hmmcopy_dir, 'metadata.yaml')
input_yaml_blob = os.path.join(hmmcopy_dir, 'input.yaml')
workflow.setobj(
obj=mgd.OutputChunks('cell_id'),
value=list(bam_files.keys()),
)
workflow.subworkflow(
name='hmmcopy_workflow',
func=hmmcopy.create_hmmcopy_workflow,
args=(mgd.InputFile('bam_markdups',
'cell_id',
fnames=bam_files,
extensions=['.bai']),
mgd.OutputFile(hmmcopy_files['reads_csvs']),
mgd.OutputFile(hmmcopy_files['segs_csvs']),
mgd.OutputFile(hmmcopy_files['metrics_csvs']),
mgd.OutputFile(hmmcopy_files['params_csvs']),
mgd.OutputFile(hmmcopy_files['igv_csvs']),
mgd.OutputFile(hmmcopy_files['segs_pdf']),
mgd.OutputFile(hmmcopy_files['bias_pdf']),
mgd.OutputFile(hmmcopy_files['heatmap_pdf']),
mgd.OutputFile(hmmcopy_files['metrics_pdf']),
mgd.OutputFile(hmmcopy_files['kernel_density_pdf']),
mgd.OutputFile(hmmcopy_files['hmmcopy_data_tar']), cellids,
config, sampleinfo),
)
workflow.transform(
name='generate_meta_files_results',
func='single_cell.utils.helpers.generate_and_upload_metadata',
args=(sys.argv[0:], hmmcopy_dir, list(hmmcopy_files.values()),
mgd.OutputFile(hmmcopy_meta)),
kwargs={
'input_yaml_data': inpututils.load_yaml(args['input_yaml']),
'input_yaml': mgd.OutputFile(input_yaml_blob),
'metadata': {
'library_id': lib,
'cell_ids': list(bam_files.keys()),
'type': 'hmmcopy',
}
})
return workflow