def process_parsed_blastoutput(dbname, blastparser, cutoffs,
annotation_results, pickorfs):
fields = [
'target', 'q_length', 'bitscore', 'bsr', 'expect', 'identity', 'ec',
'query'
]
fields.append('product')
for data in blastparser:
if data != None and isWithinCutoffs(data, cutoffs):
# if dbname=='refseq':
#if "partial" in data['product']:
# print data['query'] + '\t' + str(data['q_length']) +'\t' + str(data['bitscore']) +'\t' + str(data['expect']) +'\t' + str(data['identity']) + '\t' + str(data['bsr']) + '\t' + data['ec'] + '\t' + data['product']
annotation = {}
shortORFId = None
for field in fields:
if field in data:
if field == 'query':
shortORFId = getShortORFId(data[field])
annotation[field] = shortORFId
else:
annotation[field] = data[field]
if not shortORFId in pickorfs:
blastparser.rewind()
return None
annotation['dbname'] = dbname
if not shortORFId in annotation_results:
annotation_results[shortORFId] = []
annotation_results[shortORFId].append(annotation)
return None
def process_parsed_blastoutput(dbname, blastparser, cutoffs, annotation_results, pickorfs):
fields = ['target', 'q_length', 'bitscore', 'bsr', 'expect', 'identity', 'ec', 'query' ]
fields.append('product')
for data in blastparser:
if data!=None and isWithinCutoffs(data, cutoffs) :
# if dbname=='refseq':
#if "partial" in data['product']:
# print data['query'] + '\t' + str(data['q_length']) +'\t' + str(data['bitscore']) +'\t' + str(data['expect']) +'\t' + str(data['identity']) + '\t' + str(data['bsr']) + '\t' + data['ec'] + '\t' + data['product']
annotation = {}
shortORFId = None
for field in fields:
if field in data:
if field == 'query':
shortORFId = getShortORFId(data[field])
annotation[field] = shortORFId
else:
annotation[field] = data[field]
if not shortORFId in pickorfs:
blastparser.rewind()
return None
annotation['dbname'] = dbname
if not shortORFId in annotation_results:
annotation_results[shortORFId] = []
annotation_results[shortORFId].append(annotation)
return None
def create_annotation(results_dictionary, dbname, annotated_gff, output_dir, Taxons, orfsPicked, orfToContig, lca):
meganTree = None
#lca.set_results_dictionary(results_dictionary)
if not path.exists(output_dir):
makedirs(output_dir)
orf_dictionary={}
#process_gff_file(annotated_gff, orf_dictionary)
gffreader = GffFileParser(annotated_gff)
output_table_file = open(output_dir + '/functional_and_taxonomic_table.txt', 'a')
count = 0
for contig in gffreader:
# shortORFId = getShortORFId(orf['id'])
for orf in gffreader.orf_dictionary[contig]:
shortORFId = getShortORFId(orf['id'])
if shortORFId not in orfsPicked:
continue
orfToContig[shortORFId] = contig
taxonomy = None
#_results = re.search(r'refseq', opts_global.database_name, re.I)
if shortORFId in Taxons:
taxonomy1=Taxons[shortORFId]
taxonomy_id=lca.get_supported_taxon(taxonomy1, return_id=True)
# print taxonomy_id
preferred_taxonomy = lca.get_preferred_taxonomy(taxonomy_id)
if preferred_taxonomy:
taxonomy = preferred_taxonomy
else:
taxonomy = Taxons[shortORFId]
else:
taxonomy = 'root'
product = orf['product'] # leave product as it is
# product = re.sub(r'\[{1,2}.+?\]{1,2}','', orf['product']).strip()
# product = re.sub(r'\[[^\[]+?\]','', orf['product']).strip()
# if "partial" in orf['product']:
# print orf['product'].strip()
# print product
fprintf(output_table_file, "%s", orf['id'])
fprintf(output_table_file, "\t%s", orf['orf_length'])
fprintf(output_table_file, "\t%s", orf['start'])
fprintf(output_table_file, "\t%s", orf['end'])
fprintf(output_table_file, "\t%s", orf['seqname'])
fprintf(output_table_file, "\t%s", orf['contig_length'])
fprintf(output_table_file, "\t%s", orf['strand'])
fprintf(output_table_file, "\t%s", orf['ec'])
# fprintf(output_table_file, "\t%s", str(species))
fprintf(output_table_file, "\t%s", taxonomy)
fprintf(output_table_file, "\t%s\n", product)
# adding taxons to the megan tree
#if meganTree and taxonomy != '':
# meganTree.insertTaxon(taxonomy)
#print meganTree.getChildToParentMap()
output_table_file.close()
def compute_min_support_tree(self,
annotate_gff_file,
pickorfs,
dbname='refseq'):
self.tax_dbname = dbname
gffreader = GffFileParser(annotate_gff_file)
try:
for contig in gffreader:
for orf in gffreader.orf_dictionary[contig]:
shortORFId = getShortORFId(orf['id'])
#print shortORFId, orf['id']
if not shortORFId in pickorfs:
continue
#print ">", shortORFId, orf['id']
taxonomy = None
species = []
if self.tax_dbname in self.results_dictionary:
if shortORFId in self.results_dictionary[
self.tax_dbname]:
#compute the top hit wrt score
top_score = 0
for hit in self.results_dictionary[
self.tax_dbname][shortORFId]:
# print hit['bitscore'], self.lca_min_score, top_score
if hit['bitscore'] >= self.lca_min_score and hit[
'bitscore'] >= top_score:
top_score = hit['bitscore']
for hit in self.results_dictionary[
self.tax_dbname][shortORFId]:
if (100 - self.lca_top_percent
) * top_score / 100 < hit['bitscore']:
names = self.get_species(hit)
if names:
species.append(names)
# print self.results_dictionary[dbname][shortORFId][0]['product']
# print orf['id']
# print orf['id'], species
# print orf['id'], len(self.results_dictionary[dbname][shortORFId]), species
taxonomy = self.getTaxonomy(species)
# taxonomy_id = self.getTaxonomy(species, return_id=True)
# print taxonomy
# print taxonomy_id
# print taxonomy, orf['id'], species
self.update_taxon_support_count(taxonomy)
# preferred_taxonomy = self.get_preferred_taxonomy(taxonomy_id)
# print taxonomy
# print preferred_taxonomy
pickorfs[shortORFId] = taxonomy
except:
import traceback
traceback.print_exc()
print "ERROR : Cannot read annotated gff file "
def get_list_of_queries(annotated_gff):
orfList = {}
gffreader = GffFileParser(annotated_gff)
count = 0
for contig in gffreader:
for orf in gffreader.orf_dictionary[contig]:
orfid = getShortORFId(orf['id'])
orfList[orfid] = 1
count += 1
# if count%500000==0:
# print count
return orfList.keys()
def get_list_of_queries(annotated_gff):
orfList = {}
gffreader = GffFileParser(annotated_gff)
count = 0
for contig in gffreader:
for orf in gffreader.orf_dictionary[contig]:
orfid = getShortORFId(orf['id'])
orfList[orfid] = 1
count += 1
# if count%500000==0:
# print count
return orfList.keys()
def compute_min_support_tree(self, annotate_gff_file, pickorfs, dbname= 'refseq'):
self.tax_dbname = dbname
gffreader = GffFileParser(annotate_gff_file)
try:
for contig in gffreader:
for orf in gffreader.orf_dictionary[contig]:
shortORFId = getShortORFId(orf['id'])
#print shortORFId, orf['id']
if not shortORFId in pickorfs:
continue
#print ">", shortORFId, orf['id']
taxonomy = None
species = []
if self.tax_dbname in self.results_dictionary:
if shortORFId in self.results_dictionary[self.tax_dbname]:
#compute the top hit wrt score
top_score = 0
for hit in self.results_dictionary[self.tax_dbname][shortORFId]:
# print hit['bitscore'], self.lca_min_score, top_score
if hit['bitscore'] >= self.lca_min_score and hit['bitscore'] >= top_score:
top_score = hit['bitscore']
for hit in self.results_dictionary[self.tax_dbname][shortORFId]:
if (100-self.lca_top_percent)*top_score/100 < hit['bitscore']:
names = self.get_species(hit)
if names:
species.append(names)
# print self.results_dictionary[dbname][shortORFId][0]['product']
# print orf['id']
# print orf['id'], species
# print orf['id'], len(self.results_dictionary[dbname][shortORFId]), species
taxonomy=self.getTaxonomy(species)
# taxonomy_id = self.getTaxonomy(species, return_id=True)
# print taxonomy
# print taxonomy_id
# print taxonomy, orf['id'], species
self.update_taxon_support_count(taxonomy)
# preferred_taxonomy = self.get_preferred_taxonomy(taxonomy_id)
# print taxonomy
# print preferred_taxonomy
pickorfs[shortORFId] = taxonomy
except:
import traceback
traceback.print_exc()
print "ERROR : Cannot read annotated gff file "
def process_parsed_blastoutput(dbname, blastparser, cutoffs, annotation_results, pickorfs):
fields = ['target', 'q_length', 'bitscore', 'bsr', 'expect', 'identity', 'ec', 'query' ]
fields.append('product')
try:
for data in blastparser:
# if dbname=='refseq-nr-2014-01-18':
# print 'refseq process', data
if data!=None and isWithinCutoffs(data, cutoffs) :
#if dbname=='refseq-nr-2014-01-18':
# print 'refseq process', data
# if dbname=='refseq':
#if "partial" in data['product']:
# print data['query'] + '\t' + str(data['q_length']) +'\t' + str(data['bitscore']) +'\t' + str(data['expect']) +'\t' + str(data['identity']) + '\t' + str(data['bsr']) + '\t' + data['ec'] + '\t' + data['product']
annotation = {}
shortORFId = None
for field in fields:
if field in data:
if field == 'query':
shortORFId = getShortORFId(data[field])
annotation[field] = shortORFId
else:
annotation[field] = data[field]
if not shortORFId in pickorfs:
continue
# blastparser.rewind()
# return None
annotation['dbname'] = dbname
if not shortORFId in annotation_results:
annotation_results[shortORFId] = []
#if dbname=='refseq-nr-2014-01-18':
#print annotation
annotation_results[shortORFId].append(annotation)
except:
print traceback.print_exc()
#if dbname=='refseq-nr-2014-01-18':
# print 'annot refseq process', len(annotation_results)
return None
def merge_sorted_parsed_files(dbname,
filenames,
outputfilename,
orfRanks,
verbose=False,
errorlogger=None):
linecount = 0
readerhandles = []
if verbose:
eprintf("Processing database : %s\n", dbname)
if len(filenames) == 0:
eprintf(
"WARNING : Cannot find any B/LAST output file for database : %\n",
dbname)
exit_process()
try:
for i in range(len(filenames)):
#print filenames
readerhandles.append(BlastOutputTsvParser(dbname, filenames[i]))
except OSError:
eprintf("ERROR: Cannot read sequence file : %s\n", filenames[i])
exit_process()
# set error and warning parameters
for readerhandle in readerhandles:
readerhandle.setMaxErrorsLimit(5)
readerhandle.setErrorAndWarningLogger(errorlogger)
readerhandle.setSTEP_NAME('PARSE BLAST')
try:
outputfile = open(outputfilename, 'w')
fieldmapHeaderLine = readerhandles[0].getHeaderLine()
fprintf(outputfile, "%s\n", fieldmapHeaderLine)
except OSError:
eprintf("ERROR: Cannot create sequence file : %s\n", outputfilename)
exit_process()
values = []
for i in range(len(filenames)):
iterate = iter(readerhandles[i])
try:
next(iterate)
line = readerhandles[i].getProcessedLine()
fields = [x.strip() for x in line.split('\t')]
shortORFId = getShortORFId(fields[0])
values.append((i, orfRanks[shortORFId], line))
except:
outputfile.close()
return
S = len(filenames)
BuildHeap(S, values)
while S > 0:
try:
iterate = iter(readerhandles[values[0][0]])
line = readerhandles[values[0][0]].getProcessedLine()
fields = [x.strip() for x in line.split('\t')]
#print fields[0], orfRanks[fields[0]]
fprintf(outputfile, "%s\n", line)
next(iterate)
line = readerhandles[values[0][0]].getProcessedLine()
fields = [x.strip() for x in line.split('\t')]
shortORFId = getShortORFId(fields[0])
values[0] = (values[0][0], orfRanks[shortORFId], line)
except:
#print 'finished ' + str(S)
values[0] = values[S - 1]
S = S - 1
if S > 0:
Heapify(values, 0, S)
#print 'line count ' + str(linecount)
outputfile.close()
def create_annotation(results_dictionary, dbname, annotated_gff, output_dir,
Taxons, orfsPicked, orfToContig, lca):
meganTree = None
#lca.set_results_dictionary(results_dictionary)
if not path.exists(output_dir):
makedirs(output_dir)
orf_dictionary = {}
#process_gff_file(annotated_gff, orf_dictionary)
gffreader = GffFileParser(annotated_gff)
output_table_file = open(
output_dir + '/functional_and_taxonomic_table.txt', 'a')
count = 0
for contig in gffreader:
# shortORFId = getShortORFId(orf['id'])
for orf in gffreader.orf_dictionary[contig]:
shortORFId = getShortORFId(orf['id'])
if shortORFId not in orfsPicked:
continue
orfToContig[shortORFId] = contig
taxonomy = None
#_results = re.search(r'refseq', opts_global.database_name, re.I)
if shortORFId in Taxons:
taxonomy1 = Taxons[shortORFId]
taxonomy_id = lca.get_supported_taxon(taxonomy1,
return_id=True)
# print taxonomy_id
preferred_taxonomy = lca.get_preferred_taxonomy(taxonomy_id)
if preferred_taxonomy:
taxonomy = preferred_taxonomy
else:
taxonomy = Taxons[shortORFId]
else:
taxonomy = 'root'
product = orf['product'] # leave product as it is
# product = re.sub(r'\[{1,2}.+?\]{1,2}','', orf['product']).strip()
# product = re.sub(r'\[[^\[]+?\]','', orf['product']).strip()
# if "partial" in orf['product']:
# print orf['product'].strip()
# print product
fprintf(output_table_file, "%s", orf['id'])
fprintf(output_table_file, "\t%s", orf['orf_length'])
fprintf(output_table_file, "\t%s", orf['start'])
fprintf(output_table_file, "\t%s", orf['end'])
fprintf(output_table_file, "\t%s", orf['seqname'])
fprintf(output_table_file, "\t%s", orf['contig_length'])
fprintf(output_table_file, "\t%s", orf['strand'])
fprintf(output_table_file, "\t%s", orf['ec'])
# fprintf(output_table_file, "\t%s", str(species))
fprintf(output_table_file, "\t%s", taxonomy)
fprintf(output_table_file, "\t%s\n", product)
# adding taxons to the megan tree
#if meganTree and taxonomy != '':
# meganTree.insertTaxon(taxonomy)
#print meganTree.getChildToParentMap()
output_table_file.close()
def merge_sorted_parsed_files(dbname, filenames, outputfilename, orfRanks, verbose=False, errorlogger = None):
linecount = 0
readerhandles = []
if verbose:
eprintf("Processing for database : %s\n", dbname)
if len(filenames)==0:
eprintf("WARNING : Cannot find any B/LAST output file for database : %\n", dbname)
exit_process()
try:
for i in range(len(filenames)):
#print filenames
readerhandles.append(BlastOutputTsvParser(dbname, filenames[i]) )
except OSError:
eprintf("ERROR: Cannot read sequence file : %s\n", filenames[i])
exit_process()
# set error and warning parameters
for readerhandle in readerhandles:
readerhandle.setMaxErrorsLimit(5)
readerhandle.setErrorAndWarningLogger(errorlogger)
readerhandle.setSTEP_NAME('PARSE BLAST')
try:
outputfile = open(outputfilename, 'w')
fieldmapHeaderLine = readerhandles[0].getHeaderLine()
fprintf(outputfile, "%s\n",fieldmapHeaderLine)
except OSError:
eprintf("ERROR: Cannot create sequence file : %s\n", outputfilename)
exit_process()
values = []
for i in range(len(filenames)):
iterate = iter(readerhandles[i])
try :
next(iterate)
line = readerhandles[i].getProcessedLine()
fields = [ x.strip() for x in line.split('\t') ]
shortORFId = getShortORFId(fields[0])
values.append( (i, orfRanks[shortORFId], line) )
except:
outputfile.close()
return
S = len(filenames)
BuildHeap(S, values)
while S>0:
try:
iterate = iter(readerhandles[values[0][0]])
line = readerhandles[values[0][0]].getProcessedLine()
fields = [ x.strip() for x in line.split('\t') ]
#print fields[0], orfRanks[fields[0]]
fprintf(outputfile, "%s\n",line)
next(iterate)
line = readerhandles[values[0][0]].getProcessedLine()
fields = [ x.strip() for x in line.split('\t') ]
shortORFId = getShortORFId(fields[0])
values[0] = (values[0][0], orfRanks[shortORFId], line)
except:
#import traceback
#traceback.print_exc()
#print 'finished ' + str(S)
values[0] = values[S-1]
S = S - 1
if S>0:
Heapify(values, 0, S)
#print 'line count ' + str(linecount)
outputfile.close()