def run(self):
'It runs the analysis.'
inputs, output_dirs = self._get_inputs_and_prepare_outputs()
db_dir = output_dirs['db_dir']
pipeline = [annotate_cdna_introns]
general_settings = self._project_settings['General_settings']
settings = self._project_settings['Annotation']
genomic_seqs = settings['Cdna_intron_annotation']['genomic_seq_file']
if not genomic_seqs:
msg = 'genomic_seq_file not set for cdna intron annotation'
raise ValueError(msg)
if not settings['Cdna_intron_annotation']['genomic_db']:
project_path = general_settings['project_path']
genomic_db = make_backbone_blast_db(project_path, genomic_seqs,
dbtype='nucl')
else:
genomic_db = settings['Cdna_intron_annotation']['genomic_db']
configuration = {'annotate_cdna_introns': {'genomic_db':genomic_db,
'genomic_seqs_fhand':genomic_seqs}}
return self._run_annotation(pipeline=pipeline,
configuration=configuration,
inputs=inputs,
output_dir=db_dir)
def _get_snv_filter_specs(self):
'It gets the pipeline and configuration from settings'
configuration = {}
settings = self._project_settings['snv_filters']
for filter_data in settings.values():
if not filter_data['use']:
continue
name = filter_data['name']
unique_name = filter_data['unique_name'] if 'unique_name' in filter_data else name
filter_config = {}
for argument, value in filter_data.items():
if argument == 'use' or argument == 'name':
continue
if name == 'ref_not_in_list':
filter_config['seq_list'] = []
for item in open(value):
filter_config['seq_list'].append(item.strip())
else:
filter_config[argument] = value
if name == 'by_kind' and argument == 'kind':
filter_config[argument] = self._snv_kind_to_franklin(value)
filter_config['name'] = name
#the uniq_continguous filter can have the genomic db not defined
if (name == 'uniq_contiguous' and
('genomic_db' not in filter_config or
not filter_config['genomic_db'])):
project_path = self._project_settings['General_settings']['project_path']
genomic_db = make_backbone_blast_db(project_path,
filter_config['genomic_seqs_fpath'],
dbtype='nucl')
filter_config['genomic_db'] = genomic_db
configuration[unique_name] = filter_config
pipeline = self._get_pipeline_from_step_name(configuration)
return pipeline, configuration
