def partition_VJ(inhandle,basename):
# ignores allele numbers
def vj_id_no_allele(chain):
return seqtools.cleanup_id(chain.v.split('*')[0]) + '_' + seqtools.cleanup_id(chain.j.split('*')[0])
def outname(basename,vj_id):
return "%s.%s.imgt" % (basename,vj_id)
outhandles = {}
for chain in vdj.parse_imgt(inhandle):
curr_vj_id = vj_id_no_allele(chain)
try:
print >>outhandles[curr_vj_id], chain
except KeyError:
outhandles[curr_vj_id] = open( outname(basename,curr_vj_id), 'w' )
print >>outhandles[curr_vj_id], chain
for outhandle in outhandles.itervalues():
outhandle.close()
return [outname(basename,vj_id) for vj_id in outhandles.iterkeys()]
def partition_VJ(inhandle, basename):
# ignores allele numbers
def vj_id_no_allele(chain):
return seqtools.cleanup_id(
chain.v.split('*')[0]) + '_' + seqtools.cleanup_id(
chain.j.split('*')[0])
def outname(basename, vj_id):
return "%s.%s.imgt" % (basename, vj_id)
outhandles = {}
for chain in vdj.parse_imgt(inhandle):
curr_vj_id = vj_id_no_allele(chain)
try:
print >> outhandles[curr_vj_id], chain
except KeyError:
outhandles[curr_vj_id] = open(outname(basename, curr_vj_id), 'w')
print >> outhandles[curr_vj_id], chain
for outhandle in outhandles.itervalues():
outhandle.close()
return [outname(basename, vj_id) for vj_id in outhandles.iterkeys()]
#! /usr/bin/env python
import sys
import optparse
import vdj
import vdj.pipeline
parser = optparse.OptionParser()
(options, args) = parser.parse_args()
if len(args) == 2:
inhandle = open(args[0],'r')
outhandle = open(args[1],'w')
elif len(args) == 1:
inhandle = open(args[0],'r')
outhandle = sys.stdout
elif len(args) == 0:
inhandle = sys.stdin
outhandle = sys.stdout
for chain in vdj.parse_imgt(inhandle):
if hasattr(chain,'v') and hasattr(chain,'j'):
print >>outhandle, chain
# 1. SIZE SELECTION
log("Performing size selection on reads...")
min_size = int(params['min_size'])
max_size = int(params['max_size'])
size_selected_file = join(
work_dir, basename + '.size%i-%i' % (min_size, max_size) + '.imgt')
with open(size_selected_file, 'w') as outhandle:
for seq in SeqIO.parse(params['input_fasta'], 'fasta', generic_dna):
if len(seq) >= min_size and len(seq) <= max_size:
chain = vdj.ImmuneChain(seq)
print >> outhandle, chain
log("finished\n")
# 2. SPLIT INTO PARTS
log("Splitting input into small parts...")
parts = vdj.pipeline.iterator2parts(vdj.parse_imgt(size_selected_file),
join(work_dir, 'parts/size_selected.imgt'),
int(params['packet_size']))
log("finished\n")
# 3-7. BARCODE ID, CODING STRAND, ISOTYPE ID, VDJ CLASSIFICATION, TRANSLATION via LSF
log("Setting up LSF command...\n")
locus_options = ' '.join([' --locus %s' % locus for locus in params['locus']])
cmd = 'barcode_id.py --barcodes %s ' % params[
'barcode_fasta'] # 3. BARCODE IDENTIFICATION
cmd += ' | coding_strand.py' + locus_options # 4. CODING STRAND
if 'IGH' in params['locus']: # 5. ISOTYPE ID (heavy chain only)
cmd += ' | isotype_id.py --IGHC %s' % params['isotype_fasta']
cmd += ' | align_vdj.py' + locus_options # 6. VDJ CLASSIFICATION
cmd += ' | translate_chains.py' # 7. TRANSLATION
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import subprocess
import vdj
import vdj.analysis
# 1. Split V and J and only put in unique sequences. Sort by lineage-weighted abundance
v_counts = pyutils.nesteddict()
j_counts = pyutils.nesteddict()
for chain in vdj.parse_imgt():
v_feature_list = [chain.__getattribute__('V-REGION').qualifiers['gene'][0],chain.v_seq]
v_counts.nested_add(v_feature_list)
j_feature_list = [chain.__getattribute__('J-REGION').qualifiers['gene'][0],chain.j_seq]
j_counts.nested_add(j_feature_list)
for tup in v_counts.walk():
(keylist,val) = (tup[:-1],tup[-1])
v_counts.nested_assign(keylist,len(val))
for tup in j_counts.walk():
(keylist,val) = (tup[:-1],tup[-1])
j_counts.nested_assign(keylist,len(val))
for key in v_counts:
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import sys
import optparse
import vdj
parser = optparse.OptionParser()
(options, args) = parser.parse_args()
if len(args) == 2:
inhandle = open(args[0], 'r')
outhandle = open(args[1], 'w')
elif len(args) == 1:
inhandle = open(args[0], 'r')
outhandle = sys.stdout
elif len(args) == 0:
inhandle = sys.stdin
outhandle = sys.stdout
else:
raise Exception, "Wrong number of arguments."
for chain in vdj.parse_imgt(inhandle):
# print >>outhandle, chain.format('fasta') # causes chain.description output instead of chain.id
print >> outhandle, ">%s\n%s" % (chain.id, chain.seq)
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import sys
import optparse
import vdj
import vdj.pipeline
parser = optparse.OptionParser()
parser.add_option('-b', '--basename')
parser.add_option('-p', '--packetsize', type='int')
(options, args) = parser.parse_args()
if len(args) == 1:
inhandle = open(args[0], 'r')
elif len(args) == 0:
inhandle = sys.stdin
else:
raise Exception, "Too many arguments."
parts = vdj.pipeline.iterator2parts(vdj.parse_imgt(inhandle), options.basename,
options.packetsize)
for part in parts:
print part
#! /usr/bin/env python
import sys
import argparse
import pymongo
import vdj
import vdj.mongo
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument('-d','--db',required=True)
argparser.add_argument('-c','--collection',default='chains')
argparser.add_argument('-i','--input')
# argparser.add_argument('--option',dest='xxx',action='store_const',default=5)
args = argparser.parse_args()
inputfile = args.input
db = vdj.mongo.connect_to_spleen(connect_to=args.db)
chains = db[args.collection]
for (i,chain) in enumerate(vdj.parse_imgt(inputfile)):
if i%1000 == 0:
sys.stdout.write("%i "%i)
sys.stdout.flush()
chains.insert(vdj.mongo.encode_chain(chain))
def imgt2fasta(inhandle,outhandle):
for chain in vdj.parse_imgt(inhandle):
outhandle.write( chain.format('fasta') )
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import sys
import argparse
from collections import defaultdict
import vdj
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument('input',nargs='?',type=argparse.FileType('r'),default=sys.stdin)
argparser.add_argument('output',nargs='?',type=argparse.FileType('w'),default=sys.stdout)
args = argparser.parse_args()
# read in all the junctions
junctions = defaultdict(list)
for chain in vdj.parse_imgt(args.input):
try:
junctions[chain.junction_nt].append(chain.id)
except AttributeError:
pass
for junction in sorted(junctions.iterkeys(), key=lambda k: len(junctions[k]), reverse=True):
for id_ in junctions[junction]:
args.output.write(">%s\n%s\n" % (id_, junction))
log("Performing size selection on reads...")
min_size = int(params['min_size'])
max_size = int(params['max_size'])
size_selected_file = join(work_dir,basename + '.size%i-%i' % (min_size,max_size) + '.imgt')
with open(size_selected_file,'w') as outhandle:
for seq in SeqIO.parse(params['input_fasta'],'fasta',generic_dna):
if len(seq) >= min_size and len(seq) <= max_size:
chain = vdj.ImmuneChain(seq)
print >>outhandle, chain
log("finished\n")
# 2. SPLIT INTO PARTS
log("Splitting input into small parts...")
parts = vdj.pipeline.iterator2parts( vdj.parse_imgt(size_selected_file),
join(work_dir,'parts/size_selected.imgt'),
int(params['packet_size']))
log("finished\n")
# 3-7. BARCODE ID, CODING STRAND, ISOTYPE ID, VDJ CLASSIFICATION, TRANSLATION via LSF
log("Setting up LSF command...\n")
locus_options = ' '.join([' --locus %s' % locus for locus in params['locus']])
cmd = 'barcode_id.py --barcodes %s ' % params['barcode_fasta'] # 3. BARCODE IDENTIFICATION
cmd += ' | coding_strand.py' + locus_options # 4. CODING STRAND
if 'IGH' in params['locus']: # 5. ISOTYPE ID (heavy chain only)
cmd += ' | isotype_id.py --IGHC %s' % params['isotype_fasta']
cmd += ' | align_vdj.py' + locus_options # 6. VDJ CLASSIFICATION
cmd += ' | translate_chains.py' # 7. TRANSLATION
def imgt2fasta(inhandle, outhandle):
for chain in vdj.parse_imgt(inhandle):
outhandle.write(chain.format('fasta'))
#! /usr/bin/env python
# Copyright 2014 Uri Laserson
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import argparse
import vdj
from pyutils import cleanup_id
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument('input_file')
argparser.add_argument('output_dir',default=os.getcwd())
args = argparser.parse_args()
for chain in vdj.parse_imgt(args.input_file):
output_file = os.path.join(args.output_dir,'%s.imgt' % cleanup_id(chain.id))
with open(output_file,'w') as op:
print >>op, chain
# Copyright 2014 Uri Laserson
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import argparse
import vdj
from pyutils import cleanup_id
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument('input_file')
argparser.add_argument('output_dir', default=os.getcwd())
args = argparser.parse_args()
for chain in vdj.parse_imgt(args.input_file):
output_file = os.path.join(args.output_dir,
'%s.imgt' % cleanup_id(chain.id))
with open(output_file, 'w') as op:
print >> op, chain
def imgt2countdict(inhandle,features,count='read'):
return iterator2countdict(vdj.parse_imgt(inhandle),features,count)
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import sys
import optparse
import vdj
import vdj.pipeline
parser = optparse.OptionParser()
parser.add_option('-b','--basename')
parser.add_option('-p','--packetsize',type='int')
(options, args) = parser.parse_args()
if len(args) == 1:
inhandle = open(args[0],'r')
elif len(args) == 0:
inhandle = sys.stdin
else:
raise Exception, "Too many arguments."
parts = vdj.pipeline.iterator2parts( vdj.parse_imgt(inhandle),
options.basename,
options.packetsize)
for part in parts:
print part
(options,args) = option_parser.parse_args()
if len(args) == 2:
inhandle = open(args[0],'r')
outhandle = open(args[1],'w')
elif len(args) == 1:
inhandle = open(args[0],'r')
outhandle = sys.stdout
elif len(args) == 0:
raise ValueError, "must provide at least an input file"
# determine the total number of chains in the file (using unix grep and wc)
p = subprocess.Popen('cat %s | grep ^ID | wc -l' % args[0],shell=True,stdout=subprocess.PIPE)
total_chains = int(p.stdout.read().strip())
# check if subsampling should include the entire file
if options.num >= total_chains:
warnings.warn("Subsampling level is greater than or equal to number of chains in file: printing whole file to output.")
for chain in vdj.parse_imgt(inhandle):
print >>outhandle, chain
else:
# choose a random set of indices to select for the output file
random.seed()
idxs = sorted(random.sample(xrange(total_chains),options.num))
for (i,chain) in enumerate(vdj.parse_imgt(inhandle)):
if len(idxs) == 0:
break
if i == idxs[0]:
print >>outhandle, chain
idxs.pop(0)
def imgt2countdict(inhandle,features,count='read'):
return iterator2countdict(vdj.parse_imgt(inhandle),features,count)
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import sys
import argparse
from collections import defaultdict
import vdj
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument("input", nargs="?", type=argparse.FileType("r"), default=sys.stdin)
argparser.add_argument("output", nargs="?", type=argparse.FileType("w"), default=sys.stdout)
args = argparser.parse_args()
# read in all the junctions
junctions = defaultdict(list)
for chain in vdj.parse_imgt(args.input):
try:
junctions[chain.junction_nt].append(chain.id)
except AttributeError:
pass
for junction in sorted(junctions.iterkeys(), key=lambda k: len(junctions[k]), reverse=True):
for id_ in junctions[junction]:
args.output.write(">%s\n%s\n" % (id_, junction))
import sys
import argparse
import pymongo
import vdj
import vdj.mongo
argparser = argparse.ArgumentParser(description=None)
argparser.add_argument('-d','--db',required=True)
argparser.add_argument('-c','--collection',required=True)
argparser.add_argument('-i','--input',nargs='?',type=argparse.FileType('r'),default=sys.stdin)
argparser.add_argument('-p','--padding',type=int,default=0)
# argparser.add_argument('--option',dest='xxx',action='store_const',default=5)
args = argparser.parse_args()
inputfile = args.input
db = vdj.mongo.connect_to_lymph(connect_to=args.db)
chains = db[args.collection]
for (i,chain) in enumerate(vdj.parse_imgt(inputfile)):
if i%1000 == 0:
sys.stdout.write("%i "%i)
sys.stdout.flush()
doc = vdj.mongo.encode_chain(chain)
if args.padding > 0:
doc['__padding__'] = '0' * args.padding
chains.insert(doc, safe=True)
# delete padding from elements
if args.padding > 0:
chains.update({}, {"$unset" : {"__padding__" : 1}}, upsert=False, safe=True, multi=True)