def align(seqs):
'''
Multiple seqeuence alignment using ClustalW
@param seqs: list of biological sequences
@type seqs: list of strs
@return : aligned sequences
@rtype : list of strs
'''
name = 'tmpFoo%d' % time.time()
tfile = open(name+'.fasta','w')
try:
for i,seq in enumerate(seqs):
tfile.write('>%d\n' % (i+1))
tfile.write('%s\n' % seq)
tfile.flush()
cline = MultipleAlignCL(tfile.name)
cline.set_output(name+'.aln')
alignment = do_alignment(cline)
return [ rec.seq.tostring() for rec in alignment.get_all_seqs() ]
finally:
os.unlink(name + '.fasta')
os.unlink(name + '.dnd')
os.unlink(name + '.aln')
def align(self):
seqs = True
while seqs:
seqs = self.phamServer.request_seqs(self.server, self.numSeqs, self.client)
try:
import pynotify
if pynotify.init("Phamerator"):
n = pynotify.Notification("Phamerator Update", "doing clustalw alignments", "file:///home/steve/Applications/git/PhamDB/phageManager_logo.png")
n.show()
else:
pass
#print "there was a problem initializing the pynotify module"
except:
pass
if len(seqs) == 0:
self._logger.log('server returned no sequences to align.')
return
self._logger.log('server returned ' + str(len(seqs)) + ' to align')
self._logger.log('aligning sequences')
results = []
###########################################################################
# BEGIN MATT'S ALTERATIONS #
# #
###########################################################################
# tuple of all parallel python servers to connect with
ppservers = ()
# Creates jobserver with automatically detected number of workers
job_server = pp.Server(ppservers=ppservers)
#grab number of processors
numcpus = job_server.get_ncpus()
print "numcpus =",numcpus
#for n, person in enumerate(people):
#for seq in seqs:
for i,currentseq in enumerate(seqs):
jobs = []
id, querySeq, subjectSeq = currentseq
f = open(os.path.join(self.rootdir, 'temp' + str(i) + '.fasta'), 'w')
f.write('>' + 'a' + '\n' + querySeq + '\n>' + 'b' + '\n' + subjectSeq + '\n')
f.close()
cline = MultipleAlignCL(os.path.join(self.rootdir, 'temp' + str(i) + '.fasta'))
#cline.is_quick = True
cline.set_output(os.path.join(self.rootdir, 'temp' + str(i) + '.aln'))
#alignment = Clustalw.do_alignment(cline)
#jobs = [(input, job_server.submit(sum_primes,(input,), (isprime,), ("math",))) for input in inputs]
jobs.append(job_server.submit(clustalwAligner.run_clustalw, (id,cline), (), ("clustalwAligner.run_clustalw","Bio.Clustalw",)))
for job in jobs:
id,alignment = job()
length = alignment.get_alignment_length()
star = alignment._star_info.count('*')
score = float(star)/length
#print 'length:', length, 'identical:', star, 'score:', score
#_logger.log queryName, subjectName, score
results.append((id, score))
self._logger.log('reporting scores back to server')
self.phamServer.report_scores(results, self.server, self.client)
def Align(self, s1, s2, result):
result.clear()
handle_tmpfile1, filename_tmpfile1 = tempfile.mkstemp()
handle_tmpfile2, filename_tmpfile2 = tempfile.mkstemp()
os.write(handle_tmpfile1, ">s1\n%s\n" % (s1))
os.write(handle_tmpfile1, ">s2\n%s\n" % (s2))
os.close(handle_tmpfile1)
os.close(handle_tmpfile2)
cline = MultipleAlignCL(filename_tmpfile1)
cline.set_output(filename_tmpfile2)
align = do_alignment(cline)
seqs = align.get_all_seqs()
if len(seqs) != 2:
return result
a1 = seqs[0].seq.tostring()
a2 = seqs[1].seq.tostring()
x1 = 0
x2 = 0
for pos in range(len(a1)):
if a1[pos] not in "Nn-" and a2[pos] not in "Nn-":
result.addPair(x1, x2, 0)
x1 += 1
x2 += 1
continue
if a1[pos] != "-":
x1 += 1
if a2[pos] != "-":
x2 += 1
os.remove(filename_tmpfile1)
os.remove(filename_tmpfile2)
return result
def Align(self, s1, s2, result):
result.clear()
handle_tmpfile1, filename_tmpfile1 = tempfile.mkstemp()
handle_tmpfile2, filename_tmpfile2 = tempfile.mkstemp()
os.write(handle_tmpfile1, ">s1\n%s\n" % (s1))
os.write(handle_tmpfile1, ">s2\n%s\n" % (s2))
os.close(handle_tmpfile1)
os.close(handle_tmpfile2)
cline = MultipleAlignCL(filename_tmpfile1)
cline.set_output(filename_tmpfile2)
align = do_alignment(cline)
seqs = align.get_all_seqs()
if len(seqs) != 2:
return result
a1 = seqs[0].seq.tostring()
a2 = seqs[1].seq.tostring()
x1 = 0
x2 = 0
for pos in range(len(a1)):
if a1[pos] not in "Nn-" and a2[pos] not in "Nn-":
result.addPair(x1, x2, 0)
x1 += 1
x2 += 1
continue
if a1[pos] != "-":
x1 += 1
if a2[pos] != "-":
x2 += 1
os.remove(filename_tmpfile1)
os.remove(filename_tmpfile2)
return result
def align_seq_clustalw(sequences,mat='BLOSUM',output_order='INPUT'):
"""Return a clustal alignment of an arbitrary number of sequences. Requires clustalw on path.
Sequences are of type SeqRecord"""
assert(len(sequences) > 1)
from Bio.Clustalw import MultipleAlignCL
from Bio import Clustalw
from Bio.SeqIO.FastaIO import FastaWriter
import tempfile
def write(f):
i = 0
for seqrecord in sequences:
if seqrecord.id == "<unknown id>":
seqrecord.id = str(i)
i+=1
writer = FastaWriter(f)
writer.write_file(sequences)
f.flush() #IMPORTANT
input = tempfile.NamedTemporaryFile("w")
#input = open("input","w")
try:
write(input)
cline = MultipleAlignCL(input.name)
cline.set_protein_matrix(MultipleAlignCL.PROTEIN_MATRIX[MultipleAlignCL.PROTEIN_MATRIX.index(mat)])
cline.is_quick = False
# Set to PAM matrix.
output = tempfile.NamedTemporaryFile("w")
#output = open("output","w")
try:
cline.set_output(output.name, output_order=MultipleAlignCL.OUTPUT_ORDER[MultipleAlignCL.OUTPUT_ORDER.index(output_order)])
#print "input",open(input.name).read()
#print "cline",cline
subprocess.check_call(str(cline),shell=True,stdout=open(os.devnull,"w"))
f = open(output.name)
try:
iter = Bio.AlignIO.ClustalIO.ClustalIterator(f)
# There should only be one alignment in the file
alignments = list(iter)
assert(len(alignments)==1)
alignment = alignments[0]
finally:
f.close()
#print "output",output.read()
return alignment
finally:
output.close()
finally:
input.close()
def Align_Results(OutputFileName):
import os
FileIN_Name = """/users/rwbarrettemac/bioinformatics/pythonfolders/FMDanalysisScript/FMDserotypingARRAY/Consensus_Results/%s.FASTA""" % (OutputFileName)
FileOUT_ALN = """/users/rwbarrettemac/bioinformatics/pythonfolders/FMDanalysisScript/FMDserotypingARRAY/Consensus_Results/%s.ALN""" % (OutputFileName)
print FileIN_Name
print FileOUT_ALN
from Bio.Clustalw import MultipleAlignCL
from Bio import Clustalw
cline = MultipleAlignCL(os.path.join(os.curdir, FileIN_Name))
cline.set_output(FileOUT_ALN)
alignment = Clustalw.do_alignment(cline)
cline.close()
and parse the results into an object that can be dealt with easily."""
# standard library
import os
# biopython
from Bio.Alphabet import IUPAC
from Bio import Clustalw
from Bio.Clustalw import MultipleAlignCL
from Bio.Align import AlignInfo
from Bio.SubsMat import FreqTable
# create the command line to run clustalw
# this assumes you've got clustalw somewhere on your path, otherwise
# you need to pass a second argument to MultipleAlignCL with the complete
# path to clustalw
cline = MultipleAlignCL(os.path.join(os.curdir, 'opuntia.fasta'))
cline.set_output('test.aln')
# actually perform the alignment and get back an alignment object
alignment = Clustalw.do_alignment(cline)
# get the records in the alignment
all_records = alignment.get_all_seqs()
print 'description:', all_records[0].description
print 'sequence:', all_records[0].seq
# get the length of the alignment
print 'length', alignment.get_alignment_length()
print alignment
if "not found" not in output and "CLUSTAL" in output.upper() :
clustalw_exe = "clustalw"
if not clustalw_exe :
raise MissingExternalDependencyError(\
"Install clustalw or clustalw2 if you want to use Bio.Clustalw.")
#################################################################
print "Checking error conditions"
print "========================="
print "Empty file"
input_file = "does_not_exist.fasta"
assert not os.path.isfile(input_file)
cline = MultipleAlignCL(input_file, command=clustalw_exe)
try :
align = Clustalw.do_alignment(cline)
assert False, "Should have failed, returned %s" % repr(align)
except IOError, err :
print "Failed (good)"
#Python 2.3 on Windows gave (0, 'Error')
#Python 2.5 on Windows gives [Errno 0] Error
assert "Cannot open sequence file" in str(err) \
or "not produced" in str(err) \
or str(err) == "[Errno 0] Error" \
or str(err) == "(0, 'Error')", str(err)
print
print "Single sequence"
input_file = "Fasta/f001"
def align(self):
id = True
ppservers = ()
job_server = pp.Server(ppservers=ppservers,secret="secret")
while id: # this should test to make sure there are still alignments to do
print 'getting work unit'
try:
clustalw_work_unit = self.phamServer.request_seqs(self.client)
if not clustalw_work_unit.query_id:
print 'no work units available...sleeping'
logo = os.path.join(os.path.dirname(__file__),"pixmaps/phamerator.png")
#print "logo: %s" % logo
try:
import pynotify
if pynotify.init("Phamerator"):
n = pynotify.Notification("Phamerator Update", "No Clustalw alignments left to do...sleeping", "file:///%s" % logo)
n.show()
else:
pass
#print "there was a problem initializing the pynotify module"
except:
pass
time.sleep(30)
continue
except Exception, x:
print ''.join(Pyro.util.getPyroTraceback(x))
server, db = self.phamServer.request_db_info()
#c = db_conf(username=self.username, password=self.password, server=server, db=db)
#clustalw_work_unit.set_cursor(c)
print 'got it'
try:
import pynotify
if pynotify.init("Phamerator"):
logo = os.path.join(os.path.dirname(__file__),"pixmaps/phamerator.png")
#print "logo: %s" % logo
n = pynotify.Notification("Phamerator Update", "Clustalw alignments in progress for id %s" % clustalw_work_unit.query_id, "file:///%s" % logo)
n.show()
else:
pass
#print "there was a problem initializing the pynotify module"
except:
pass
self._logger.log('aligning sequences')
###########################################################################
# BEGIN MATT'S ALTERATIONS #
# #
###########################################################################
results = []
open_files = []
# tuple of all parallel python servers to connect with
# Creates jobserver with automatically detected number of workers
#grab number of processors
numcpus = job_server.get_ncpus()
print "numcpus =",numcpus
#for n, person in enumerate(people):
#for seq in seqs:
jobs = []
#for i,currentseq in enumerate(seqs):
query_id = clustalw_work_unit.query_id
query_translation = clustalw_work_unit.query_translation
counter = 0
for record in clustalw_work_unit.database:
subject_id, subject_translation = record.id, record.translation
fname = os.path.join(self.rootdir, 'temp' + query_id + '_' + subject_id + '.fasta')
f = open(fname, 'w')
open_files.append(fname)
open_files.append(fname.replace('.fasta','.dnd'))
open_files.append(fname.replace('.fasta','.aln'))
f.write('>%s\n%s\n>%s\n%s\n' % (query_id, query_translation, subject_id, subject_translation))
f.close()
clustalw_infile = os.path.join(self.rootdir, 'temp' + str(query_id) + '_' + str(subject_id) + '.fasta')
if float(Bio.__version__) >= 1.56:
# pass the query id (qid) and the subject id (sid) to run_clustalw
jobs.append(job_server.submit(clustalwAligner.run_clustalw, (clustalw_infile, query_id, subject_id), (), ()))
else:
cline = MultipleAlignCL(clustalw_infile)
cline.set_output(os.path.join(self.rootdir, 'temp' + str(query_id) + '_' + str(subject_id) + '.aln'))
# pass the query id (qid) and the subject id (sid) to run_clustalw
jobs.append(job_server.submit(clustalwAligner.run_clustalw_old, (query_id, subject_id,cline), (), ("Bio.Clustalw",)))
counter = counter + 3
if counter > 50:
results = self.process_jobs(jobs, results, open_files)
jobs = []
open_files = []
counter = 0
results = self.process_jobs(jobs, results, open_files)
jobs = []
open_files = []
counter = 0
# must report everything back in atomic transaction
print 'reporting scores back to server'
try:
self.phamServer.report_scores(clustalw_work_unit, results, self.client)
except Exception, x:
print ''.join(Pyro.util.getPyroTraceback(x))
print 'exiting on pyro traceback'
sys.exit()
and "Multiple Sequence Alignments" in output:
clustalw_exe = "clustalw"
if not clustalw_exe:
raise MissingExternalDependencyError(\
"Install clustalw or clustalw2 if you want to use Bio.Clustalw.")
#################################################################
print "Checking error conditions"
print "========================="
print "Empty file"
input_file = "does_not_exist.fasta"
assert not os.path.isfile(input_file)
cline = MultipleAlignCL(input_file, command=clustalw_exe)
try:
align = Clustalw.do_alignment(cline)
assert False, "Should have failed, returned %s" % repr(align)
except IOError, err:
print "Failed (good)"
#Python 2.3 on Windows gave (0, 'Error')
#Python 2.5 on Windows gives [Errno 0] Error
assert "Cannot open sequence file" in str(err) \
or "not produced" in str(err) \
or str(err) == "[Errno 0] Error" \
or str(err) == "(0, 'Error')", str(err)
print
print "Single sequence"
input_file = "Fasta/f001"
def align(self):
id = True
ppservers = ()
job_server = pp.Server(ppservers=ppservers, secret="secret")
while id: # this should test to make sure there are still alignments to do
print 'getting work unit'
try:
clustalw_work_unit = self.phamServer.request_seqs(self.client)
if not clustalw_work_unit.query_id:
print 'no work units available...sleeping'
logo = os.path.join(os.path.dirname(__file__),
"pixmaps/phamerator.png")
#print "logo: %s" % logo
try:
import pynotify
if pynotify.init("Phamerator"):
n = pynotify.Notification(
"Phamerator Update",
"No Clustalw alignments left to do...sleeping",
"file:///%s" % logo)
n.show()
else:
pass
#print "there was a problem initializing the pynotify module"
except:
pass
time.sleep(30)
continue
except Exception, x:
print ''.join(Pyro.util.getPyroTraceback(x))
server, db = self.phamServer.request_db_info()
#c = db_conf(username=self.username, password=self.password, server=server, db=db)
#clustalw_work_unit.set_cursor(c)
print 'got it'
try:
import pynotify
if pynotify.init("Phamerator"):
logo = os.path.join(os.path.dirname(__file__),
"pixmaps/phamerator.png")
#print "logo: %s" % logo
n = pynotify.Notification(
"Phamerator Update",
"Clustalw alignments in progress for id %s" %
clustalw_work_unit.query_id, "file:///%s" % logo)
n.show()
else:
pass
#print "there was a problem initializing the pynotify module"
except:
pass
self._logger.log('aligning sequences')
###########################################################################
# BEGIN MATT'S ALTERATIONS #
# #
###########################################################################
results = []
open_files = []
# tuple of all parallel python servers to connect with
# Creates jobserver with automatically detected number of workers
#grab number of processors
numcpus = job_server.get_ncpus()
print "numcpus =", numcpus
#for n, person in enumerate(people):
#for seq in seqs:
jobs = []
#for i,currentseq in enumerate(seqs):
query_id = clustalw_work_unit.query_id
query_translation = clustalw_work_unit.query_translation
counter = 0
for record in clustalw_work_unit.database:
subject_id, subject_translation = record.id, record.translation
fname = os.path.join(
self.rootdir,
'temp' + query_id + '_' + subject_id + '.fasta')
f = open(fname, 'w')
open_files.append(fname)
open_files.append(fname.replace('.fasta', '.dnd'))
open_files.append(fname.replace('.fasta', '.aln'))
f.write('>%s\n%s\n>%s\n%s\n' %
(query_id, query_translation, subject_id,
subject_translation))
f.close()
clustalw_infile = os.path.join(
self.rootdir,
'temp' + str(query_id) + '_' + str(subject_id) + '.fasta')
if float(Bio.__version__) >= 1.56:
# pass the query id (qid) and the subject id (sid) to run_clustalw
jobs.append(
job_server.submit(
clustalwAligner.run_clustalw,
(clustalw_infile, query_id, subject_id), (), ()))
else:
cline = MultipleAlignCL(clustalw_infile)
cline.set_output(
os.path.join(
self.rootdir, 'temp' + str(query_id) + '_' +
str(subject_id) + '.aln'))
# pass the query id (qid) and the subject id (sid) to run_clustalw
jobs.append(
job_server.submit(clustalwAligner.run_clustalw_old,
(query_id, subject_id, cline), (),
("Bio.Clustalw", )))
counter = counter + 3
if counter > 50:
results = self.process_jobs(jobs, results, open_files)
jobs = []
open_files = []
counter = 0
results = self.process_jobs(jobs, results, open_files)
jobs = []
open_files = []
counter = 0
# must report everything back in atomic transaction
print 'reporting scores back to server'
try:
self.phamServer.report_scores(clustalw_work_unit, results,
self.client)
except Exception, x:
print ''.join(Pyro.util.getPyroTraceback(x))
print 'exiting on pyro traceback'
sys.exit()
and parse the results into an object that can be dealt with easily."""
# standard library
import os
# biopython
from Bio.Alphabet import IUPAC
from Bio import Clustalw
from Bio.Clustalw import MultipleAlignCL
from Bio.Align import AlignInfo
from Bio.SubsMat import FreqTable
# create the command line to run clustalw
# this assumes you've got clustalw somewhere on your path, otherwise
# you need to pass a second argument to MultipleAlignCL with the complete
# path to clustalw
cline = MultipleAlignCL(os.path.join(os.curdir, "opuntia.fasta"))
cline.set_output("test.aln")
# actually perform the alignment and get back an alignment object
alignment = Clustalw.do_alignment(cline)
print alignment
print "first description:", alignment[0].description
print "first sequence:", alignment[0].seq
# get the length of the alignment
print "length", alignment.get_alignment_length()
print alignment
