def blastall_file2db(fname,dbname="",blastprogram="blastp",output="ncbiparsed",extra_blastp_params={'F': 'F', 'e': '10'}):
"""
"""
if blastprogram not in ['blastp','tblastn','blastn','tblastx']:
raise "only blastp and tblastn are supported"
extra_params = " ".join(["-%s %s" % (k,v) for k,v in extra_blastp_params.iteritems()])
command = "%s -p %s %s -i %s -d %s " % (BLASTALL_PATH,blastprogram,extra_params,fname,dbname)
try:
ci,co,ce = osPopen3(command)
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
# do NOT remove the input fname
except:
co.close()
error = ce.read().strip()
ce.close()
print command
print "ERROR: '%s'" % error
raise "BLAST CRASHED...."
# and return!
return blastallout
def run_cexpander_dr(settingsfile, outfname, verbose=False):
"""
Run the cexpander_dr program by specifying input & output filenames etc.
@type infname: string
@param infname: (absolute) path to cbalignp output file
@type outfname: string
@param outfname: (absolute) path to cexpander_dr output file
@type commandline: string
@param commandline: literal additional command line arguments for cexpander_dr
@type verbose: Boolean
@param verbose: print debugging messages to STDOUT (True) or not (False, default)
@attention: requires global variable EXECUTABLE_CEXPANDER_CEXPANDER
@attention: see cexpander_dr for (additional) command line options
@attention: only a subset of cexpander_dr commandline options are supported!
@rtype: Boolean
@return: True
"""
command = "%s < %s > %s " % (EXECUTABLE_CEXPANDER_CEXPANDER, settingsfile,
outfname)
if verbose: print command
ci, co, ce = osPopen3(command)
ci.close()
co.close()
ce.close()
return True
def run_cexpander_dr(settingsfile,outfname,verbose=False):
"""
Run the cexpander_dr program by specifying input & output filenames etc.
@type infname: string
@param infname: (absolute) path to cbalignp output file
@type outfname: string
@param outfname: (absolute) path to cexpander_dr output file
@type commandline: string
@param commandline: literal additional command line arguments for cexpander_dr
@type verbose: Boolean
@param verbose: print debugging messages to STDOUT (True) or not (False, default)
@attention: requires global variable EXECUTABLE_CEXPANDER_CEXPANDER
@attention: see cexpander_dr for (additional) command line options
@attention: only a subset of cexpander_dr commandline options are supported!
@rtype: Boolean
@return: True
"""
command = "%s < %s > %s " % (
EXECUTABLE_CEXPANDER_CEXPANDER,
settingsfile,
outfname
)
if verbose: print command
ci,co,ce = osPopen3(command)
ci.close()
co.close()
ce.close()
return True
def blastall_seq2seq(fastadata=(),filenames=(),output="ncbiparsed",blastprogram="blastp",remove_files=True,extra_blastp_params={'F': 'F', 'e': '10'}):
"""
choose proper input:
fastadata ( ( headerQUERY, seqQUERY ) , ( headerSBJCT, seqSBJCT ) )
or
filenames ( filenameQUERY, filenameSBJCT )
"""
input = None
if blastprogram not in ['blastp','tblastn','tblastx','blastx']:
raise "only blastp and tblastn are supported"
elif blastprogram in ['tblastn','tblastx']:
dna_or_prot = "F"
else:
dna_or_prot = "T"
if fastadata and type(fastadata) == type(()) and len(fastadata) == 2 and not filenames:
# input is fasta headers and sequence
input = "fastadata"
# write input filenames
uniquetag = get_random_string_tag()
fname_q = "_".join( [ uniquetag, str(fastadata[0][0]), 'Q.fa' ] )
fname_s = "_".join( [ uniquetag, str(fastadata[1][0]), 'S.fa' ] )
fh = open(fname_q,'w')
fh.write(">%s\n%s" % (fastadata[0][0],fastadata[0][1]))
fh.close()
fh = open(fname_s,'w')
fh.write(">%s\n%s" % (fastadata[1][0],fastadata[1][1]))
fh.close()
elif filenames and type(filenames) == type(()) and len(filenames) == 2 and not fastadata:
# input is (supposed to be) filenames
input = "filenames"
# get filenames
fname_q = filenames[0]
fname_s = filenames[1]
elif not filenames and not fastadata:
raise "no input!"
else:
raise "inproper input!"
# formatdb
OSsystem("%s -i %s -p %s" % (FORMATDB_PATH,fname_s,dna_or_prot))
# and blastall!
extra_params = " ".join(["-%s %s" % (k,v) for k,v in extra_blastp_params.iteritems()])
ci,co,ce = osPopen3("%s -p %s %s -i %s -d %s " % (BLASTALL_PATH,blastprogram,extra_params,fname_q,fname_s))
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
if remove_files:
OSsystem("rm %s.*" % fname_s)
osRemove("%s" % fname_s)
osRemove("%s" % fname_q)
# and return!
return blastallout
def blastall_seq2db(header,
sequence,
dbname="",
blastprogram="blastp",
output="ncbiparsed",
extra_blastp_params={
'F': 'F',
'e': '10'
}):
"""
"""
if blastprogram not in ['blastp', 'tblastn', 'blastn', 'blastx']:
raise "only blastp and tblastn are supported"
extra_params = " ".join(
["-%s %s" % (k, v) for k, v in extra_blastp_params.iteritems()])
# generate (semi ;-) unique filename
uniquetag = get_random_string_tag()
fname = "_".join(
[uniquetag,
str(header).replace(" ", "_"), sequence[0:10] + ".fa"])
fname = osPathJoin(OSgetcwd(), fname)
fh = open(fname, 'w')
fh.write(">%s\n%s\n" % (header, sequence))
fh.close()
command = "%s -p %s %s -i %s -d %s " % (BLASTALL_PATH, blastprogram,
extra_params, fname, dbname)
try:
ci, co, ce = osPopen3(command)
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
except:
# for some kind of - obvious or freak accident case -
# Blast or parsing of the blast record failed
# No debugging here; just cleanup and return False
print "BLAST CRASHED::"
print command
blastallout = False
# remove the created Query file
osRemove(fname)
# and return!
return blastallout
def getorf(sequence=None,fname=None,outputfile=None,
executable_getorf=EXECUTABLE_GETORF,
minsize=EXECUTABLE_GETORF_MINSIZE):
"""
Run EMBOSS getorf and write results to file
@type sequence: string
@param inputfile: full path to getorf input file (fasta)
@type outputfile: string
@param outputfile: full path to getorf output file (fasta)
@type executable_getorf: string
@param executable_getorf: full path to getorf executable
"""
# do some integrity checks
if not sequence and not fname:
message = "specify `sequence` or `fname` variable, not neither"
raise InproperlyAppliedArgument, message
if sequence and fname:
message = "specify `sequence` or `fname` variable, not both"
raise InproperlyAppliedArgument, message
if not executable_getorf:
message = "specify `EXECUTABLE_GETORF` variable"
raise InproperlyAppliedArgument, message
# create command line, execute with popen and parse
command = "%s -minsize %s -noreverse" % (
executable_getorf,
minsize
)
if fname: command = "cat %s | %s" % (fname, command)
else: command = "echo %s | %s" % (sequence, command)
if outputfile: command = "%s -outseq %s" % (command, outputfile)
else: command = "%s -filter" % (command)
ci,co,ce = osPopen3(command)
ci.close()
output = co.read()
co.close()
error = ce.read()
ce.close()
if error and error.strip() != "Finds and extracts open reading frames (ORFs)":
print "getorf ERROR:", error
if not outputfile:
return output
else:
return outputfile
def run_getorf(inputfile="",outputfile="",
executable_getorf=EXECUTABLE_GETORF,
minsize=EXECUTABLE_GETORF_MINSIZE):
"""
Run EMBOSS getorf and write results to file
@type inputfile: string
@param inputfile: full path to getorf input file (fasta)
@type outputfile: string
@param outputfile: full path to getorf output file (fasta)
@type EXECUTABLE_GETORF: string
@param EXECUTABLE_GETORF: full path to getorf executable
"""
# do some integrity checks
if not inputfile:
raise InproperlyAppliedArgument, "specify `inputfile` variable"
if not outputfile:
raise InproperlyAppliedArgument, "specify `outputfile` variable"
if not executable_getorf:
raise InproperlyAppliedArgument, "specify `executable_getorf` variable"
# create command line, execute with popen and parse
command = "%s -sequence %s -outseq %s -minsize %s -noreverse" % (
executable_getorf,
inputfile,
outputfile,
minsize
)
ci,co,ce = osPopen3(command)
ci.close()
output = co.read()
co.close()
error = ce.read()
ce.close()
if output:
print "getorf OUTPUT:", output
if error and error.strip() != "Finds and extracts open reading frames (ORFs)":
print "getorf ERROR:", error
def blastall_file2db(fname,
dbname="",
blastprogram="blastp",
output="ncbiparsed",
extra_blastp_params={
'F': 'F',
'e': '10'
}):
"""
"""
if blastprogram not in ['blastp', 'tblastn', 'blastn', 'tblastx']:
raise "only blastp and tblastn are supported"
extra_params = " ".join(
["-%s %s" % (k, v) for k, v in extra_blastp_params.iteritems()])
command = "%s -p %s %s -i %s -d %s " % (BLASTALL_PATH, blastprogram,
extra_params, fname, dbname)
try:
ci, co, ce = osPopen3(command)
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
# do NOT remove the input fname
except:
co.close()
error = ce.read().strip()
ce.close()
print command
print "ERROR: '%s'" % error
raise "BLAST CRASHED...."
# and return!
return blastallout
def blastall_seq2db(header,sequence,dbname="",blastprogram="blastp",output="ncbiparsed",extra_blastp_params={'F': 'F', 'e': '10'}):
"""
"""
if blastprogram not in ['blastp','tblastn','blastn','blastx']:
raise "only blastp and tblastn are supported"
extra_params = " ".join(["-%s %s" % (k,v) for k,v in extra_blastp_params.iteritems()])
# generate (semi ;-) unique filename
uniquetag = get_random_string_tag()
fname = "_".join( [ uniquetag, str(header).replace(" ","_"), sequence[0:10]+".fa" ] )
fname = osPathJoin(OSgetcwd(),fname)
fh = open(fname,'w')
fh.write(">%s\n%s\n" % (header,sequence))
fh.close()
command = "%s -p %s %s -i %s -d %s " % (BLASTALL_PATH,blastprogram,extra_params,fname,dbname)
try:
ci,co,ce = osPopen3(command)
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
except:
# for some kind of - obvious or freak accident case -
# Blast or parsing of the blast record failed
# No debugging here; just cleanup and return False
print "BLAST CRASHED::"
print command
blastallout = False
# remove the created Query file
osRemove(fname)
# and return!
return blastallout
def runcexpander(fname_fasta, cbalignp_commandline=" -y", output='binary'):
"""
Run the complete cascade of cexpander algorithms on an input multi fasta
file and return the output as a CexpanderOutput object
@type fname_fasta: string
@param fname_fasta: path to input multi fasta file
@type cbalignp_commandline: string
@param cbalignp_commandline: (extra) command line for cbalignp
@type min_cols: integer
@param min_cols: minimal number of uniformly matched positions (cols)
required to report transfer blocks for (>= 0)
@type projected_on: string
@param projected_on: apply fasta seqeunce header which to use for projection;
apply ':::' to do projections on all input sequences
@attention: requires global variable EXECUTABLE_cexpander_ALLVSALL
@attention: requires global variable EXECUTABLE_CEXPANDER_CBALIGNP
@attention: requires global variable EXECUTABLE_CEXPANDER_CEXPANDER
@attention: see cexpander_dr for (additional) command line options
@attention: only a subset of cexpander_dr commandline options are supported!
@rtype: CexpanderOutput object
@return: CexpanderOutput object
"""
if not fname_fasta: raise "NoProperFunctionArguments"
if not osPathIsfile(fname_fasta): raise "FileDoesNotExist"
# (0) create (~unique) filenames
uniquetag = get_random_string_tag()
fname_allvsall = ".".join([fname_fasta, uniquetag, "allvsall"])
fname_report = ".".join([fname_fasta, uniquetag, "report"])
fname_aligned = ".".join([fname_fasta, uniquetag, "aligned"])
fname_settings = ".".join([fname_fasta, uniquetag, "settings"])
fname_cexpander = ".".join([fname_fasta, uniquetag, "cexpander"])
# (1) create complete .fa -> cexpanderstring command
command = """
python %s %s %s %s;
%s -i %s %s > %s;
%s < %s;
""" % (
EXECUTABLE_CEXPANDER_ALLVSALL,
fname_fasta,
fname_allvsall,
fname_report,
EXECUTABLE_CEXPANDER_CBALIGNP,
fname_allvsall,
cbalignp_commandline,
fname_aligned,
EXECUTABLE_CEXPANDER_CEXPANDER,
fname_settings,
)
# (2) create fname_settings file
binorfloat = "$dumpcv"
if output == "float": binorfloat = "$dumpcvc"
fh = open(fname_settings, 'w')
content = "\n\n".join([
"$load\n%s\n%s" % (fname_report, fname_aligned),
"$addquery\n-1",
"$run",
"$dumpentries",
"$cv_linear",
"%s" % (binorfloat), # BINARY == $dumpcv, FLOAT = $dumpcvc
"$exit\n\n",
])
fh.write(content)
fh.close()
# (3) run the command
ci, co, ce = osPopen3(command)
ci.close()
# output of EXECUTABLE_CEXPANDER_ALLVSALL is cast to STDOUT as well!
cexpanderdata = co.read()
co.close()
error = ce.read()
ce.close()
# (4) parse fname_cexpander to CexpanderOutput object
cxpdr = parse_cexpander(cexpanderdata, fname_fasta)
# (5) cleanup files
osSystem("rm -f %s %s.%s.*" % (fname_fasta, fname_fasta, uniquetag))
# (6) return the output object
return cxpdr
def blastall_seq2seq(fastadata=(),
filenames=(),
output="ncbiparsed",
blastprogram="blastp",
remove_files=True,
extra_blastp_params={
'F': 'F',
'e': '10'
}):
"""
choose proper input:
fastadata ( ( headerQUERY, seqQUERY ) , ( headerSBJCT, seqSBJCT ) )
or
filenames ( filenameQUERY, filenameSBJCT )
"""
input = None
if blastprogram not in ['blastp', 'tblastn', 'tblastx', 'blastx']:
raise "only blastp and tblastn are supported"
elif blastprogram in ['tblastn', 'tblastx']:
dna_or_prot = "F"
else:
dna_or_prot = "T"
if fastadata and type(fastadata) == type(
()) and len(fastadata) == 2 and not filenames:
# input is fasta headers and sequence
input = "fastadata"
# write input filenames
uniquetag = get_random_string_tag()
fname_q = "_".join([uniquetag, str(fastadata[0][0]), 'Q.fa'])
fname_s = "_".join([uniquetag, str(fastadata[1][0]), 'S.fa'])
fh = open(fname_q, 'w')
fh.write(">%s\n%s" % (fastadata[0][0], fastadata[0][1]))
fh.close()
fh = open(fname_s, 'w')
fh.write(">%s\n%s" % (fastadata[1][0], fastadata[1][1]))
fh.close()
elif filenames and type(filenames) == type(
()) and len(filenames) == 2 and not fastadata:
# input is (supposed to be) filenames
input = "filenames"
# get filenames
fname_q = filenames[0]
fname_s = filenames[1]
elif not filenames and not fastadata:
raise "no input!"
else:
raise "inproper input!"
# formatdb
OSsystem("%s -i %s -p %s" % (FORMATDB_PATH, fname_s, dna_or_prot))
# and blastall!
extra_params = " ".join(
["-%s %s" % (k, v) for k, v in extra_blastp_params.iteritems()])
ci, co, ce = osPopen3(
"%s -p %s %s -i %s -d %s " %
(BLASTALL_PATH, blastprogram, extra_params, fname_q, fname_s))
ci.close()
if output == "ncbiparsed":
b_parser = NCBIStandalone.BlastParser()
blastallout = b_parser.parse(co)
else:
blastallout = co.read()
co.close()
ce.close()
if remove_files:
OSsystem("rm %s.*" % fname_s)
osRemove("%s" % fname_s)
osRemove("%s" % fname_q)
# and return!
return blastallout
def runcexpander(fname_fasta,cbalignp_commandline=" -y",output='binary'):
"""
Run the complete cascade of cexpander algorithms on an input multi fasta
file and return the output as a CexpanderOutput object
@type fname_fasta: string
@param fname_fasta: path to input multi fasta file
@type cbalignp_commandline: string
@param cbalignp_commandline: (extra) command line for cbalignp
@type min_cols: integer
@param min_cols: minimal number of uniformly matched positions (cols)
required to report transfer blocks for (>= 0)
@type projected_on: string
@param projected_on: apply fasta seqeunce header which to use for projection;
apply ':::' to do projections on all input sequences
@attention: requires global variable EXECUTABLE_cexpander_ALLVSALL
@attention: requires global variable EXECUTABLE_CEXPANDER_CBALIGNP
@attention: requires global variable EXECUTABLE_CEXPANDER_CEXPANDER
@attention: see cexpander_dr for (additional) command line options
@attention: only a subset of cexpander_dr commandline options are supported!
@rtype: CexpanderOutput object
@return: CexpanderOutput object
"""
if not fname_fasta: raise "NoProperFunctionArguments"
if not osPathIsfile(fname_fasta): raise "FileDoesNotExist"
# (0) create (~unique) filenames
uniquetag = get_random_string_tag()
fname_allvsall = ".".join([fname_fasta,uniquetag,"allvsall"])
fname_report = ".".join([fname_fasta,uniquetag,"report"])
fname_aligned = ".".join([fname_fasta,uniquetag,"aligned"])
fname_settings = ".".join([fname_fasta,uniquetag,"settings"])
fname_cexpander = ".".join([fname_fasta,uniquetag,"cexpander"])
# (1) create complete .fa -> cexpanderstring command
command = """
python %s %s %s %s;
%s -i %s %s > %s;
%s < %s;
""" % (
EXECUTABLE_CEXPANDER_ALLVSALL,
fname_fasta,
fname_allvsall,
fname_report,
EXECUTABLE_CEXPANDER_CBALIGNP,
fname_allvsall,
cbalignp_commandline,
fname_aligned,
EXECUTABLE_CEXPANDER_CEXPANDER,
fname_settings,
)
# (2) create fname_settings file
binorfloat = "$dumpcv"
if output == "float": binorfloat = "$dumpcvc"
fh = open(fname_settings,'w')
content = "\n\n".join( [
"$load\n%s\n%s" % (fname_report,fname_aligned),
"$addquery\n-1",
"$run",
"$dumpentries",
"$cv_linear",
"%s" % ( binorfloat ), # BINARY == $dumpcv, FLOAT = $dumpcvc
"$exit\n\n",
] )
fh.write(content)
fh.close()
# (3) run the command
ci,co,ce = osPopen3(command)
ci.close()
# output of EXECUTABLE_CEXPANDER_ALLVSALL is cast to STDOUT as well!
cexpanderdata = co.read()
co.close()
error = ce.read()
ce.close()
# (4) parse fname_cexpander to CexpanderOutput object
cxpdr = parse_cexpander(cexpanderdata,fname_fasta)
# (5) cleanup files
osSystem("rm -f %s %s.%s.*" % ( fname_fasta, fname_fasta,uniquetag ) )
# (6) return the output object
return cxpdr