def write_annotation_for_orf(outputgff_file, candidatedbname, dbname_weight,
results_dictionary, orf_dictionary, contig,
candidate_orf_pos, orfid, compact_output):
try:
fields = [
'source', 'feature', 'start', 'end', 'score', 'strand', 'frame'
]
output_line = orf_dictionary[contig][candidate_orf_pos]['seqname']
#if compact_output:
output_line = ShortenContigId(output_line)
for field in fields:
# printf("\t%s", orf_dictionary[contig][candidate_orf_pos][field])
output_line += "\t" + str(
orf_dictionary[contig][candidate_orf_pos][field])
#if compact_output:
try:
attributes = "ID=" + ShortenORFId(
orf_dictionary[contig][candidate_orf_pos]['id'])
attributes += ";" + "locus_tag=" + ShortenORFId(
orf_dictionary[contig][candidate_orf_pos]['locus_tag'])
except:
attributes = "ID=" + orf_dictionary[contig][candidate_orf_pos]['id']
attributes += ";" + "locus_tag=" + orf_dictionary[contig][
candidate_orf_pos]['locus_tag']
attributes += ";" + "contig_length=" + orf_dictionary[contig][
candidate_orf_pos]['contig_length']
attributes += ";" + "orf_length=" + orf_dictionary[contig][
candidate_orf_pos]['orf_length']
attributes += ";" + "partial=" + orf_dictionary[contig][
candidate_orf_pos]['partial']
attributes += ";" + "sourcedb=" + candidatedbname
if candidatedbname in results_dictionary:
attributes += ";" + "annotvalue=" + str(
results_dictionary[candidatedbname][orfid]['value'])
attributes += ";" + "ec=" + str(
results_dictionary[candidatedbname][orfid]['ec'])
attributes += ";" + "product=" + results_dictionary[
candidatedbname][orfid]['product']
else:
attributes += ";" + "annotvalue=" + str('0')
attributes += ";" + "ec=" + str('')
attributes += ";" + "product=" + 'hypothetical protein'
output_line += '\t' + attributes
if candidatedbname in results_dictionary:
fprintf(outputgff_file, "%s\n", output_line)
except:
eprintf("ERROR : Failure to annotate in contig %s\n", contig)
#print orf_dictionary[contig]
print traceback.print_exc(10)
exit_process()
def insert_orf_into_dict(line, contig_dict, shortenorfid=False):
rawfields = re.split('\t', line)
fields = []
for field in rawfields:
fields.append(field.strip());
if( len(fields) != 9):
return
attributes = {}
seqname = fields[0]
try:
if shortenorfid:
seqname = ShortenContigId(fields[0])
else:
seqname = fields[0]
except:
seqname = fields[0]
attributes['seqname'] = seqname # this is a bit of a duplication
attributes['source'] = fields[1]
attributes['feature'] = fields[2]
attributes['start'] = int(fields[3])
attributes['end'] = int(fields[4])
try:
attributes['score'] = float(fields[5])
except:
attributes['score'] = fields[5]
attributes['strand'] = fields[6]
attributes['frame'] = fields[7]
split_attributes(fields[8], attributes)
if not seqname in contig_dict :
contig_dict[seqname] = []
contig_dict[seqname].append(attributes)
def create_annotation(results_dictionary, dbname, annotated_gff, output_dir, Taxons, orfsPicked, orfToContig, lca, compact_output= False, sample_name = ""):
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_name = output_dir + PATHDELIM + sample_name + ".functional_and_taxonomic_table.txt"
output_table_file = open(output_table_name, 'a')
count = 0
for contig in gffreader:
# shortORFId = getShortORFId(orf['id'])
for orf in gffreader.orf_dictionary[contig]:
shortORFId = getShortORFId(orf['id'])
count +=1
#shortORFId = ShortenORFId(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]
#print taxonomy1, 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
orf_id = orf['id']
seqname = orf['seqname']
if compact_output:
orf_id = ShortenORFId(orf_id)
seqname = ShortenContigId(seqname)
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", 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 print_orf_table(results, orfToContig, output_dir, outputfile, compact_output=False):
addHeader =True
if not path.exists(output_dir):
makedirs(output_dir)
orf_dict = {}
for dbname in results.keys():
print dbname, len(results[dbname].keys())
for orfname in results[dbname]:
for orf in results[dbname][orfname]:
if not orf['query'] in orf_dict:
orf_dict[orf['query']] = {}
if dbname in orf_dict[orf['query']]: # only the best hit prevails
continue
#if orf['query']=='2_0' and dbname=='refseq-nr-2014-01-18':
# print orf
orf_dict[orf['query']]['contig'] = orfToContig[orfname]
product = orf['product'].strip()
_results = re.search(r'cog', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = cog_id(product)
continue
_results = re.search(r'kegg', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = kegg_id(product)
continue
_results = re.search(r'cazy', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = cazy_id(product)
continue
_results = re.search(r'metacyc', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = product
continue
_results = re.search(r'seed', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = seed_id(product)
# print "---", orf_dict[orf['query']][dbname]
continue
#if dbname=='refseq-nr-2014-01-18':
# if orf['query']=='2_0':
# print product
#adds it anyway
orf_dict[orf['query']][dbname] = product
# compute the databases
database_maps = {}
for dbname in results.keys():
_results = re.search(r'cog', dbname, re.I)
if _results:
database_maps['cog'] = dbname
continue
_results = re.search(r'kegg', dbname, re.I)
if _results:
database_maps['kegg'] = dbname
hit=True
continue
_results = re.search(r'cazy', dbname, re.I)
if _results:
database_maps['cazy'] = dbname
continue
_results = re.search(r'seed', dbname, re.I)
if _results:
database_maps['seed'] = dbname
continue
_results = re.search(r'metacyc', dbname, re.I)
if _results:
database_maps['metacyc'] = dbname
continue
_results = re.search(r'refseq', dbname, re.I)
if _results:
database_maps['refseq'] = dbname
continue
database_maps[dbname] = dbname
std_dbnames = ['cog', 'kegg', 'seed', 'cazy', 'metacyc', 'refseq']
dbnames = std_dbnames
headers = ["# ORF_ID", "CONTIG_ID"]
for std_dbname in std_dbnames:
headers.append(std_dbname.upper())
for dbname in sorted(results.keys()):
non_std =True
for std_dbname in std_dbnames:
if re.search(std_dbname, dbname, re.I):
non_std =False
if non_std:
dbnames.append(dbname)
headers.append(std_dbname)
sampleName = None
for orfn in orf_dict:
#if orfn=='2_0':
# print orfn, '<<', orf_dict[orfn], ' >> xxxx'
#_keys = orf_dict[orfn].keys()
#_results = re.search(r'cog', dbname, re.I)
if 'cog' in database_maps and database_maps['cog'] in orf_dict[orfn]:
cogFn = orf_dict[orfn][database_maps['cog']]
else:
cogFn = ""
if 'kegg' in database_maps and database_maps['kegg'] in orf_dict[orfn]:
keggFn = orf_dict[orfn][database_maps['kegg']]
#print orfn, keggFn
else:
keggFn = ""
if 'metacyc' in database_maps and database_maps['metacyc'] in orf_dict[orfn]:
metacycPwy = orf_dict[orfn][database_maps['metacyc']]
else:
metacycPwy = ""
if 'seed' in database_maps and database_maps['seed'] in orf_dict[orfn]:
seedFn = orf_dict[orfn][database_maps['seed']]
else:
seedFn = ""
if 'cazy' in database_maps and database_maps['cazy'] in orf_dict[orfn]:
cazyFn = orf_dict[orfn][database_maps['cazy']]
else:
cazyFn= ""
if 'refseq' in database_maps and database_maps['refseq'] in orf_dict[orfn]:
refseqFn = orf_dict[orfn][database_maps['refseq']]
else:
refseqFn= ""
if not sampleName:
sampleName = getSampleNameFromContig(orf_dict[orfn]['contig'])
orfName = orfn
contigName= orf_dict[orfn]['contig']
if compact_output:
orfName = orfn
contigName= ShortenContigId(contigName)
row = [ orfName, contigName ]
for dbname in dbnames:
if dbname in database_maps and database_maps[dbname] in orf_dict[orfn]:
row.append(orf_dict[orfn][database_maps[dbname]])
else:
row.append("")
# print '\t'.join(headers)
# print '\t'.join(row)
#fprintf(outputfile, "%s\n", orfName + "\t" + contigName + '\t' + cogFn + '\t' + keggFn +'\t' + seedFn + '\t' + cazyFn + '\t'+ metacycPwy)
if addHeader:
#fprintf(outputfile, "# %s\n", '\t'.join(headers)_"ORF_ID" + "\t" + "CONTIG_ID" + '\t' + "COG" + '\t' + "KEGG" +'\t' + "SEED" + '\t' + "CAZY" + '\t'+ "METACYC" + '\t' + "REFSEQ" )
fprintf(outputfile, "# %s\n", '\t'.join(headers))
addHeader=False
#fprintf(outputfile, "%s\n", orfName + "\t" + contigName + '\t' + cogFn + '\t' + keggFn +'\t' + seedFn + '\t' + cazyFn + '\t'+ metacycPwy + '\t' + refseqFn )
fprintf(outputfile, "%s\n", '\t'.join(row))
def print_orf_table(results, orfToContig, output_dir, outputfile, compact_output=False):
if not path.exists(output_dir):
makedirs(output_dir)
orf_dict = {}
for dbname in results.keys():
print dbname, len(results[dbname].keys())
for orfname in results[dbname]:
for orf in results[dbname][orfname]:
if not orf['query'] in orf_dict:
orf_dict[orf['query']] = {}
if dbname in orf_dict[orf['query']]:
continue
#if orf['query']=='2_0' and dbname=='refseq-nr-2014-01-18':
# print orf
orf_dict[orf['query']]['contig'] = orfToContig[orfname]
product = orf['product'].strip()
_results = re.search(r'cog', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = cog_id(product)
continue
_results = re.search(r'kegg', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = kegg_id(product)
continue
_results = re.search(r'cazy', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = cazy_id(product)
continue
_results = re.search(r'metacyc', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = product
continue
_results = re.search(r'seed', dbname, re.I)
if _results:
orf_dict[orf['query']][dbname] = seed_id(product)
# print "---", orf_dict[orf['query']][dbname]
continue
#if dbname=='refseq-nr-2014-01-18':
# if orf['query']=='2_0':
# print product
orf_dict[orf['query']][dbname] = product
# compute the databases
database_maps = {}
for dbname in results.keys():
_results = re.search(r'cog', dbname, re.I)
if _results:
database_maps['cog'] = dbname
_results = re.search(r'kegg', dbname, re.I)
if _results:
database_maps['kegg'] = dbname
_results = re.search(r'cazy', dbname, re.I)
if _results:
database_maps['cazy'] = dbname
_results = re.search(r'seed', dbname, re.I)
if _results:
database_maps['seed'] = dbname
_results = re.search(r'metacyc', dbname, re.I)
if _results:
database_maps['metacyc'] = dbname
_results = re.search(r'refseq', dbname, re.I)
if _results:
database_maps['refseq'] = dbname
sampleName = None
for orfn in orf_dict:
#if orfn=='2_0':
# print orfn, '<<', orf_dict[orfn], ' >> xxxx'
#_keys = orf_dict[orfn].keys()
#_results = re.search(r'cog', dbname, re.I)
if 'cog' in database_maps and database_maps['cog'] in orf_dict[orfn]:
cogFn = orf_dict[orfn][database_maps['cog']]
else:
cogFn = ""
if 'kegg' in database_maps and database_maps['kegg'] in orf_dict[orfn]:
keggFn = orf_dict[orfn][database_maps['kegg']]
#print orfn, keggFn
else:
keggFn = ""
if 'metacyc' in database_maps and database_maps['metacyc'] in orf_dict[orfn]:
metacycPwy = orf_dict[orfn][database_maps['metacyc']]
else:
metacycPwy = ""
if 'seed' in database_maps and database_maps['seed'] in orf_dict[orfn]:
seedFn = orf_dict[orfn][database_maps['seed']]
else:
seedFn = ""
if 'cazy' in database_maps and database_maps['cazy'] in orf_dict[orfn]:
cazyFn = orf_dict[orfn][database_maps['cazy']]
else:
cazyFn= ""
if 'refseq' in database_maps and database_maps['refseq'] in orf_dict[orfn]:
refseqFn = orf_dict[orfn][database_maps['refseq']]
else:
refseqFn= ""
if not sampleName:
sampleName = getSampleNameFromContig(orf_dict[orfn]['contig'])
orfName = orfn
contigName= orf_dict[orfn]['contig']
if compact_output:
orfName = orfn
contigName= ShortenContigId(contigName)
#fprintf(outputfile, "%s\n", orfName + "\t" + contigName + '\t' + cogFn + '\t' + keggFn +'\t' + seedFn + '\t' + cazyFn + '\t'+ metacycPwy)
fprintf(outputfile, "%s\n", orfName + "\t" + contigName + '\t' + cogFn + '\t' + keggFn +'\t' + seedFn + '\t' + cazyFn + '\t'+ metacycPwy + '\t' + refseqFn )