def test_blastdb_seq_kind():
'It test the blastdb kind'
blastdb = join(TEST_DATA_DIR, 'blast', 'tomato_genome2')
assert guess_blastdb_kind(blastdb) == 'nucl'
blastdb = join(TEST_DATA_DIR, 'blast', 'tair7_pep')
assert guess_blastdb_kind(blastdb) == 'prot'
def _get_b2g_blast(self, input_fpath, goblast_settings):
'It gets a chopped blast ready for use with blast2go'
if 'kind' in goblast_settings:
db_kind = goblast_settings['kind']
else:
db_kind = guess_blastdb_kind(goblast_settings['path'])
seq_type = scrape_info_from_fname(input_fpath)['st']
blast_program = guess_blast_program(seq_type, db_kind,
prefer_tblastx=True)
blastdb = goblast_settings['path']
project_dir = self._project_settings['General_settings']['project_path']
blast = backbone_blast_runner(query_fpath=input_fpath,
project_dir=project_dir,
blast_program=blast_program,
blast_db=blastdb,
dbtype=db_kind,
threads=self.threads)
chop_big_xml, num_items = True, 2
if chop_big_xml:
#chopped_blast = open('/tmp/blast_itemized.xml', 'w')
chopped_blast = NamedTemporaryFile(suffix='.xml')
for blast_parts in xml_itemize(blast, 'Iteration', num_items):
chopped_blast.write(blast_parts)
chopped_blast.flush()
return chopped_blast
else:
return open(blast)
def run(self):
'It runs the analysis'
inputs, output_dirs = self._get_inputs_and_prepare_outputs()
db_dir = output_dirs['db_dir']
blast_settings = self._project_settings['blast']
settings = self._project_settings['Annotation']
annot_settings = settings['description_annotation']
description_databases = annot_settings['description_databases']
general_settings = self._project_settings['General_settings']
#first we need some blasts
project_dir = general_settings['project_path']
blasts = {}
for input_ in inputs['input']:
input_fpath = input_.last_version
for database in description_databases:
if 'kind' in blast_settings[database]:
db_kind = blast_settings[database]['kind']
else:
db_kind = guess_blastdb_kind(blast_settings[database]['path'])
seq_type = scrape_info_from_fname(input_.last_version)['st']
blast_program = guess_blast_program(seq_type, db_kind,
prefer_tblastx=True)
blastdb = blast_settings[database]['path']
blast = backbone_blast_runner(query_fpath=input_fpath,
project_dir=project_dir,
blast_program=blast_program,
blast_db=blastdb,
dbtype=db_kind,
threads=self.threads)
if input_ not in blasts:
blasts[input_fpath] = []
blasts[input_fpath].append({'blast':blast, 'modifier':None})
#print blasts
pipeline = []
configuration = {}
for database in description_databases:
step = annotate_with_descriptions
step['name_in_config'] = database
pipeline.append(step)
for input_ in inputs['input']:
step_config = {'blasts': blasts[input_.last_version]}
configuration[input_.basename] = {}
configuration[input_.basename][database] = step_config
#print configuration
return self._run_annotation(pipeline=pipeline,
configuration=configuration,
inputs=inputs,
output_dir=db_dir)
def run(self):
'It runs the analysis'
inputs, output_dirs = self._get_inputs_and_prepare_outputs()
output_dir = output_dirs['result']
blast_settings = self._project_settings['blast']
settings = self._project_settings['Annotation']['ortholog_annotation']
ortholog_databases = settings['ortholog_databases']
general_settings = self._project_settings['General_settings']
project_dir = general_settings['project_path']
#first we need some blasts
blasts = {}
for input_ in inputs['input']:
for database in ortholog_databases:
if 'kind' in blast_settings[database]:
db_kind = blast_settings[database]['kind']
else:
db_kind = guess_blastdb_kind(blast_settings[database]['path'])
seq_type = scrape_info_from_fname(input_.last_version)['st']
blast_program = guess_blast_program(seq_type, db_kind,
prefer_tblastx=True)
blastdb = blast_settings[database]['path']
if 'subj_def_as_acc' in blast_settings[database]:
subj_def_as_acc = blast_settings[database]['subj_def_as_acc']
else:
subj_def_as_acc = None
#this could be different adding something to the settings
blastdb_seq_fpath = blastdb
blast = backbone_blast_runner(query_fpath=input_.last_version,
project_dir=project_dir,
blast_program=blast_program,
blast_db=blastdb,
dbtype=db_kind,
threads=self.threads)
blast = {'fpath':blast,
'subj_def_as_acc': subj_def_as_acc}
blast_program = guess_blast_program(db_kind, seq_type,
prefer_tblastx=True)
reverse_blast = backbone_blast_runner(
query_fpath=blastdb_seq_fpath,
project_dir=project_dir,
blast_program=blast_program,
blast_db_seq=input_.last_version,
dbtype='nucl',
threads=self.threads)
reverse_blast = {'fpath':reverse_blast,
'subj_def_as_acc':None}
if input_ not in blasts:
blasts[input_] = {}
blasts[input_][database] = {'blast':blast,
'reverse_blast':reverse_blast}
pipeline = []
configuration = {}
for database in ortholog_databases:
step = copy.deepcopy(annotate_orthologs)
step['name_in_config'] = database
#an annotation step for every ortholog database
pipeline.append(step)
for input_ in inputs['input']:
reverse_blast = ''
step_config = {
'blast':{'blast': blasts[input_][database]['blast']['fpath'],
'subj_def_as_acc':blasts[input_][database]['blast']['subj_def_as_acc']},
'reverse_blast':{'blast':
blasts[input_][database]['reverse_blast']['fpath'],
'subj_def_as_acc':blasts[input_][database]['reverse_blast']['subj_def_as_acc']},
'species': database}
if input_.basename not in configuration:
configuration[input_.basename] = {}
configuration[input_.basename][database] = step_config
return self._run_annotation(pipeline=pipeline,
configuration=configuration,
inputs=inputs,
output_dir=output_dir)