class AlignmentTargetGenerator(object):
'''
Class used to retrieve the list of all possible targets from certain alignment
'''
def __init__(self):
self.crawler = DirectoryCrawler()
self.description_parser = DescriptionParser()
def get_blastn_targets(self, protein_id):
'''
@param protein_id: retrieves the list of species (RBS) not aligned with blastn for that protein
'''
path = self.crawler.get_blastn_path(protein_id)
return get_species_list(protein_id, path)
def set_failed_blastn_targets(self, protein_id, failed_species_list):
path = self.crawler.get_blastn_path(protein_id)
write_failed_species_to_status(failed_species_list, path)
def get_tblastn_targets(self, protein_id):
'''
@param protein_id: retrieves the list of species not aligned with tblastn for that protein
'''
path = self.crawler.get_tblastn_path(protein_id)
return get_species_list(protein_id, path)
def set_failed_tblastn_targets(self, protein_id, failed_species_list):
path = self.crawler.get_tblastn_path(protein_id)
write_failed_species_to_status(failed_species_list, path)
def get_SW_gene_targets(self, protein_id):
'''
@param protein_id: retrieves the list of species not aligned with SW_gene for that protein
'''
path = self.crawler.get_SW_gene_path(protein_id)
return get_species_list(protein_id, path)
def set_failed_SW_gene_targets(self, protein_id, failed_species_list):
path = self.crawler.get_SW_gene_path(protein_id)
write_failed_species_to_status(failed_species_list, path)
def get_SW_exon_targets(self, protein_id):
'''
@param protein_id: retrieves the list of species not aligned with SW_exon for that protein
'''
path = self.crawler.get_SW_exon_path(protein_id)
return get_species_list(protein_id, path)
def set_failed_SW_exon_targets(self, protein_id, failed_species_list):
path = self.crawler.get_SW_exon_path(protein_id)
write_failed_species_to_status(failed_species_list, path)
def reset_action (protein_id, key):
update_entry_in_status_file(protein_id, key, 'FAILED')
crawler = DirectoryCrawler()
if key == 'GENE_RETRIEVAL':
clear_directory(crawler.get_gene_path(protein_id))
elif key == 'EXP_GENE_RETRIEVAL' :
clear_directory(crawler.get_expanded_gene_path(protein_id))
elif key == 'PROTEIN_RETRIEVAL' :
clear_directory(crawler.get_protein_path(protein_id))
elif key == 'ENSEMBL_EXON_RETRIEVAL' :
clear_directory(crawler.get_exon_ensembl_path(protein_id))
elif key == 'GENEWISE_EXON_RETRIEVAL' :
clear_directory(crawler.get_exon_genewise_path(protein_id))
clear_directory(crawler.get_genewise_path(protein_id))
elif key == 'REF_SP_DB_FORMATTING' :
clear_directory(crawler.get_database_path(protein_id))
elif key == 'BLASTN_ALIGNMENT' :
clear_directory(crawler.get_blastn_path(protein_id))
elif key == 'TBLASTN_ALIGNMENT' :
clear_directory(crawler.get_tblastn_path(protein_id))
elif key == 'SW_GENE_ALIGNMENT' :
clear_directory(crawler.get_SW_gene_path(protein_id))
elif key == 'SW_EXON_ALIGNMENT' :
clear_directory(crawler.get_SW_exon_path(protein_id))
def generate_tblastn_alignments(protein_id, species_list = None, referenced_species = "Homo_sapiens"):
'''
Runs the tblastn program for a specified protein and list of species
@param protein_id
@param species_list: if provided, runs tblastn for this list of species, \
otherwise runs for species that are missing the tblastn output \
who are determined by .status file in the tblastn folder.
'''
logger = Logger.Instance()
alignment_logger = logger.get_logger('alignment')
alignment_generator = AlignmentTargetGenerator()
crawler = DirectoryCrawler()
command_generator = CommandGenerator()
if (not species_list):
species_list = alignment_generator.get_tblastn_targets(protein_id)
failed_species_list = []
for species in species_list:
############## MOVE
output_file = "{0}/{1}.blastout".format(crawler.get_tblastn_path(protein_id), species.strip())
input_file = "{0}/{1}.fa".format(crawler.get_protein_path(protein_id), species.strip())
database = "{0}/{1}.fa".format(crawler.get_database_path(protein_id), referenced_species)
command = command_generator.generate_tblastn_command(database, input_file, output_file)
command_return = Popen(command, shell=True, stdin=PIPE, stdout=PIPE, stderr=STDOUT, close_fds=True)
output = command_return.stdout.read()
if output != "":
#LOGGING
os.remove(output_file)
alignment_logger.warning("{0}, {1}, TBLASTN, {2}".format(protein_id, species.strip(), output.strip()))
failed_species_list.append(species.strip())
if failed_species_list:
alignment_generator.set_failed_tblastn_targets(protein_id, failed_species_list)
return False
return True
def parse_blast_output (ref_protein_id, species, blast):
'''
@return: Dictionary where key is reference species exon_id, and
the value is list of corresponding alignments
'''
logger = Logger.Instance()
containers_logger = logger.get_logger('containers')
dc = DirectoryCrawler()
if blast == "blastn":
blast_file = "{0}/{1}.blastout".format(dc.get_blastn_path(ref_protein_id), species)
else:
blast_file = "{0}/{1}.blastout".format(dc.get_tblastn_path(ref_protein_id), species)
if not os.path.isfile(blast_file):
containers_logger.error ("{0}, {1}, {2}, no blastout file".format(ref_protein_id, species, blast))
return None
file_handle = open(blast_file, 'r')
# parse blastn output
try:
blastn_record = NCBIXML.read(file_handle)
except ValueError:
containers_logger.error("%s,%s,%s,No hits found" % (ref_protein_id, species, blast))
return None
exon_dict = {}
exon_pattern = re.compile("(\d+)\|(\d+)\|(ENS\w+)\|(ENS\w+)\|([-]*1)")
for alignment in blastn_record.alignments:
(blast_info, exon_info) = alignment.title.split()
pattern_match = re.match(exon_pattern, exon_info)
ref_exon_id = pattern_match.groups()[3]
exon_start = int (pattern_match.groups()[0])
exon_end = int(pattern_match.groups()[1])
# limit alignments to 10 hsps
num_of_hsps = 0
for hsp in alignment.hsps:
# limit!
if blast == "blastn":
(query_frame, hit_frame) = hsp.frame
if query_frame == -1 or hit_frame == -1:
continue
if num_of_hsps == 5:
break
num_of_hsps += 1
exon = Exon(blast, ref_exon_id, ref_protein_id, species)
if type(hsp.gaps) is int:
gaps = hsp.gaps
elif type(hsp.gaps) is tuple:
if not hsp.gaps[0]:
gaps = 0
exon.set_alignment_info ( hsp.identities,
hsp.positives,
gaps,
hsp.sbjct_start,
hsp.sbjct_start + len(hsp.sbjct) -1,
hsp.query_start,
hsp.query_start + len(hsp.sbjct) -1,
len(hsp.sbjct),
hsp.sbjct,
hsp.query,
hsp.score)
if not ref_exon_id in exon_dict:
exon_dict[ref_exon_id] = [exon]
else:
exon_dict[ref_exon_id].append(exon)
# means we covered the whole exon
if len(hsp.sbjct) == abs(exon_end-exon_start)+1 and len(hsp.sbjct) == hsp.identities:
break
file_handle.close()
return exon_dict
