class AlignmentCommandGenerator(object):
'''
Generates commands for utilities that are used (blast, sw, genewise, fastacmd, formatdb)
'''
def __init__(self, user=None, passwd=None, host=None, port =None, check=True):
'''
Loads the utils configuration (utils.cfg)
if (not os.path.isfile("../utils.cfg")):
raise IOError("There is no utils.cfg file present in the project directory.")
config = ConfigParser.RawConfigParser()
config.read("../utils.cfg")
'''
self.configReader = ConfigurationReader(user, passwd, host, port, check)
# blast tools
self.fastacmd = self.configReader.get_path('fastacmd')
self.blastall = self.configReader.get_path('mafft')
self.blastp = self.configReader.get_path('blastall')
self.blastp += " -p blastp -e "+ self.configReader.get_value('blastp_e_value')
# ensembl database
self.ensembldb = self.configReader.get_path('ensembl_fasta')
# Smith-Waterman
self.sw_sharp = self.configReader.get_path('sw#')
# hacked blast, to be used by swsharp to align exons while respecting the boundaries
self.blosum_matrix = "{0}/{1}".format(self.configReader.get_path('resources'),
self.configReader.get_value('blosum_hacked'))
# mafft
self.mafft = self.configReader.get_path('mafft')
def generate_fastacmd_gene_command (self, species, seq_name, fasta_db_file,
strand = None,
sequence_start = None, sequence_stop = None,
output_file_path = None):
if (isinteger(seq_name)):
seq_id_cmd = "-s 'lcl|%s' " % seq_name
else:
seq_id_cmd = "-s %s" % seq_name
data_type_cmd = "-p F"
database = "-d {0}/{1}/dna/{2}".format(self.ensembldb, species, fasta_db_file)
if (strand == None or int(strand) == 1):
strand_cmd = "-S 1"
else:
strand_cmd = "-S 2"
if (sequence_start and sequence_stop):
location_cmd = "-L %s,%s" % (sequence_start, sequence_stop)
else:
location_cmd = ""
if output_file_path:
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
fastacmd_cmd_line = self.fastacmd
fastacmd_cmd_line += " {0} {1} {2} {3} {4} {5}".format(database, seq_id_cmd,
data_type_cmd, strand_cmd,
location_cmd, output_cmd)
return fastacmd_cmd_line
def generate_fastacmd_plain (self, database, seq_id, output_file_path = ''):
database_cmd = "-d %s" % database
if (isinteger(seq_id)):
seq_id_cmd = "-s 'lcl|%s' " % seq_id
else:
seq_id_cmd = "-s %s" % seq_id
if (output_file_path):
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
fastacmd_cmd_line = self.fastacmd
fastacmd_cmd_line += " {0} {1} {2} ".format(database_cmd, seq_id_cmd, output_cmd)
return fastacmd_cmd_line
def generate_fastacmd_protein_command (self, protein_id, species_name, protein_type, output_file_path):
data_type_cmd = "-p T"
prot_id_cmd = "-s %s" % protein_id
database = "-d %s" % self._generate_proteindb_file_name(species_name, protein_type)
if (output_file_path):
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
return "fastacmd {0} {1} {2} {3}".format(prot_id_cmd, data_type_cmd, database, output_cmd)
def generate_blastp_plain (self, database, input_file, output_file):
# sombbody hardcoded the output format in the self.blastp
blastp = self.blastp + " -m 8 "
if output_file:
cmd = "{0} -d {1} -i {2} -o {3}".format(blastp, database, input_file, output_file)
else:
cmd = "{0} -d {1} -i {2} ".format(blastp, database, input_file)
return cmd
def generate_blastp_command (self, database, input_file, output_file):
if output_file:
cmd = "{0} -d {1} -i {2} -o {3}".format(self.blastp, database, input_file, output_file)
else:
cmd = "{0} -d {1} -i {2} ".format(self.blastp, database, input_file)
return cmd
def generate_SW_nt (self, query_sequence_file, target_fasta_db_file,
output_file, supress_stdout = True):
# Matija's current implementation is switching the order
cmd = "{0} -j {1} -i {2} --out {3}".format(self.sw_sharp, query_sequence_file,
target_fasta_db_file, output_file)
if supress_stdout:
cmd += " > /dev/null"
return cmd
def generate_SW_peptide (self, query_sequence_file, target_fasta_db_file,
output_file = None):
cmd = "{0} --verbose 0 --matrix-file {1} ".format(self.sw_sharp, self.blosum_matrix)
cmd += " -j {0} -i {1} ".format(query_sequence_file, target_fasta_db_file)
cmd += " --out-type 1 --gap-open 3.0 "
if output_file:
cmd += " --out {0} ".format(output_file)
return cmd
def generate_formatdb_command (self, input_db_file, sequence_type):
if sequence_type == "protein" or sequence_type == "P":
cmd = "formatdb -i {0} -p T".format(input_db_file)
else:
cmd = "formatdb -i {0} -p F".format(input_db_file)
return cmd
def generate_mafft_command (self, input_file, output_file=None):
'''
If there are unusual characters (e.g., U as selenocysteine in protein sequence),
use the --anysymbol option. It accepts any printable characters (U, O, #, $, %, etc.;
0x21-0x7e in the ASCII code), execpt for > (0x3e).
They are scored equivalently to X. Gap is - (0x2d), as in the default mode.
'''
if output_file:
return "{0} --quiet --anysymbol {1} > {2}".format(self.mafft, input_file, output_file)
else:
return "{0} --quiet --anysymbol {1} ".format(self.mafft, input_file)
def _generate_genedb_file_name (self, species, sequence_type, sequence_id, masked):
'''
@param species: species name
@param sequence_type: scaffold / chromosome...
@param sequence_id: ensembl sequence ID
@param masked: 0 if dna should not be masked, 1 if it should
'''
file_name = "{0}/{1}/dna".format(self.ensembldb, species.lower())
# get the template name (dependent on the assembly)
tmp_file=""
for f in os.listdir(file_name):
# check if we've stumbled on the protein file
m = re.findall(".dna", f)
if not m:
continue
tmp_file = f
m = re.findall ('(.*).dna', tmp_file)
if (masked != 0):
file_name = "%s/%s.dna_rm." % (file_name, m[0])
else :
file_name = "%s/%s.dna." % (file_name, m[0])
if (sequence_type == 'chromosome'):
file_name = "%schromosome.%s.fa" % (file_name, sequence_id)
else :
file_name = "%stoplevel.fa" % (file_name)
if (not os.path.exists(file_name)):
# find *dna.toplevel.fa file
path = "{0}/{1}/dna".format(self.ensembldb, species.lower())
cmd = "ls "+path+"/*dna.toplevel.fa"
retval = commands.getoutput(cmd)
if (retval):
file_name = retval.rstrip()
# otherwise just let the thing fail, I have no better idea
return file_name
def _generate_proteindb_file_name (self, species, protein_type):
'''
@param species: species name (ensembl)
@param protein_type: all / abinitio
@return: protein database name
'''
file_name = "%s/%s/pep" % (self.ensembldb, species.lower())
tmp_file=""
for f in os.listdir(file_name):
# check if we've stumbled on the protein file
m = re.findall(".pep", f)
if not m:
continue
tmp_file = f
m = re.findall ('(.*).pep', tmp_file)
if (protein_type == "all"):
file_name = "%s/%s.pep.all.fa" % (file_name, m[0])
else:
file_name = "%s/%s.pep.abinitio.fa" % (file_name, m[0])
return file_name
class AlignmentCommandGenerator(object):
"""
Generates commands for utilities that are used (blast, sw, genewise, fastacmd, formatdb)
"""
def _check_exists(self, item):
if not os.path.exists(item):
print item, " not found "
exit(1)
def __init__(self, user=None, passwd=None, host=None, port=None, check=True):
"""
Loads the utils configuration (utils.cfg)
if (not os.path.isfile("../utils.cfg")):
raise IOError("There is no utils.cfg file present in the project directory.")
config = ConfigParser.RawConfigParser()
config.read("../utils.cfg")
"""
self.configReader = ConfigurationReader(user, passwd, host, port, check)
# blast tools
self.fastacmd = self.configReader.get_path("fastacmd")
self._check_exists(self.fastacmd)
self.blastall = self.configReader.get_path("blastall")
self._check_exists(self.blastall)
self.blastp = self.configReader.get_path("blastall")
self.blastp += " -p blastp -e " + str(self.configReader.get_value("blastp_e_value"))
# ensembl database
self.ensembldb = self.configReader.get_path("ensembl_fasta")
self._check_exists(self.ensembldb)
# Smith-Waterman
self.sw_sharp = self.configReader.get_path("sw#")
self._check_exists(self.sw_sharp)
# usearch
self.usearch = self.configReader.get_path("usearch")
self._check_exists(self.usearch)
# hacked blast, to be used by swsharp to align exons while respecting the boundaries
self.blosum_matrix = "{0}/{1}".format(
self.configReader.get_path("resources"), self.configReader.get_value("blosum_hacked")
)
# mafft
self.mafft = self.configReader.get_path("mafft")
self._check_exists(self.mafft)
# maxent scan
self.scan3 = self.configReader.get_path("score3")
self.scan5 = self.configReader.get_path("score5")
self.maxent_homedir = self.configReader.get_path("maxentscan")
def generate_fastacmd_gene_command(
self,
species,
seq_name,
fasta_db_file,
strand=None,
sequence_start=None,
sequence_stop=None,
output_file_path=None,
):
if seq_name.isdigit():
seq_id_cmd = "-s 'lcl|%s' " % seq_name
else:
seq_id_cmd = "-s %s" % seq_name
data_type_cmd = "-p F"
database = "-d {0}/{1}/dna/{2}".format(self.ensembldb, species, fasta_db_file)
if strand == None or int(strand) == 1:
strand_cmd = "-S 1"
else:
strand_cmd = "-S 2"
if not sequence_start and not sequence_stop:
location_cmd = ""
else:
location_cmd = "-L %s,%s" % (
sequence_start if sequence_start else "",
sequence_stop if sequence_stop else "",
)
if output_file_path:
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
fastacmd_cmd_line = self.fastacmd
fastacmd_cmd_line += " {0} {1} {2} {3} {4} {5}".format(
database, seq_id_cmd, data_type_cmd, strand_cmd, location_cmd, output_cmd
)
return fastacmd_cmd_line
def generate_fastacmd_plain(self, database, seq_id, output_file_path=""):
database_cmd = "-d %s" % database
if isinteger(seq_id):
seq_id_cmd = "-s 'lcl|%s' " % seq_id
else:
seq_id_cmd = "-s %s" % seq_id
if output_file_path:
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
fastacmd_cmd_line = self.fastacmd
fastacmd_cmd_line += " {0} {1} {2} ".format(database_cmd, seq_id_cmd, output_cmd)
return fastacmd_cmd_line
def generate_fastacmd_protein_command(self, protein_id, species_name, protein_type, output_file_path):
data_type_cmd = "-p T"
prot_id_cmd = "-s %s" % protein_id
database = "-d %s" % self._generate_proteindb_file_name(species_name, protein_type)
if output_file_path:
output_cmd = "-o %s" % output_file_path
else:
output_cmd = ""
return "fastacmd {0} {1} {2} {3}".format(prot_id_cmd, data_type_cmd, database, output_cmd)
def generate_blastp_plain(self, database, input_file, output_file):
# sombbody hardcoded the output format in the self.blastp
blastp = self.blastp + " -m 8 "
if output_file:
cmd = "{0} -d {1} -i {2} -o {3}".format(blastp, database, input_file, output_file)
else:
cmd = "{0} -d {1} -i {2} ".format(blastp, database, input_file)
return cmd
def generate_blastp_command(self, database, input_file, output_file):
if output_file:
cmd = "{0} -d {1} -i {2} -o {3}".format(self.blastp, database, input_file, output_file)
else:
cmd = "{0} -d {1} -i {2} ".format(self.blastp, database, input_file)
return cmd
def generate_usearch_nt(
self, query_sequence_file, target_fasta_db_file, output_file, identity="0.5", strand="plus"
):
# strand can also be both
cmd = "{0} -threads 1 -search_local {1} -db {2} -alnout {3} -id {4} -strand {5} ".format(
self.usearch, query_sequence_file, target_fasta_db_file, output_file, identity, strand
)
return cmd
def generate_SW_nt(self, query_sequence_file, target_fasta_db_file, output_file=None, supress_stdout=False):
# Matija's current implementation is switching the order
cmd = "{0} --verbose 0 -j {1} -i {2}".format(self.sw_sharp, query_sequence_file, target_fasta_db_file)
if not output_file == None:
cmd += " --out " + output_file
if supress_stdout:
cmd += " > /dev/null"
return cmd
def generate_SW_peptide(self, query_sequence_file, target_fasta_db_file, output_file=None):
cmd = "{0} --verbose 0 --matrix-file {1} ".format(self.sw_sharp, self.blosum_matrix)
cmd += " -j {0} -i {1} ".format(query_sequence_file, target_fasta_db_file)
cmd += " --out-type 1 --gap-open 3.0 "
if output_file:
cmd += " --out {0} ".format(output_file)
return cmd
def generate_formatdb_command(self, input_db_file, sequence_type):
if sequence_type == "protein" or sequence_type == "P":
cmd = "formatdb -i {0} -p T".format(input_db_file)
else:
cmd = "formatdb -i {0} -p F".format(input_db_file)
return cmd
def generate_mafft_command(self, input_file, output_file=None):
"""
If there are unusual characters (e.g., U as selenocysteine in protein sequence),
use the --anysymbol option. It accepts any printable characters (U, O, #, $, %, etc.;
0x21-0x7e in the ASCII code), execpt for > (0x3e).
They are scored equivalently to X. Gap is - (0x2d), as in the default mode.
"""
if output_file:
return "{0} --quiet --anysymbol {1} > {2}".format(self.mafft, input_file, output_file)
else:
return "{0} --quiet --anysymbol {1} ".format(self.mafft, input_file)
def generate_mafft_profile(self, input_file_1, input_file_2, output_file=None):
if output_file:
return "{0}-profile {1} {2} > {3}".format(self.mafft, input_file_1, input_file_2, output_file)
else:
return "{0}-profile {1} {2} ".format(self.mafft, input_file_1, input_file_2)
def generate_maxentscan_cmd(self, intron_side, input_file):
if intron_side == 3:
return "{0} {1} {2} ".format(self.scan3, self.maxent_homedir, input_file)
elif intron_side == 5:
return "{0} {1} {2} ".format(self.scan5, self.maxent_homedir, input_file)
else:
return ""
def _generate_genedb_file_name(self, species, sequence_type, sequence_id, masked):
"""
@param species: species name
@param sequence_type: scaffold / chromosome...
@param sequence_id: ensembl sequence ID
@param masked: 0 if dna should not be masked, 1 if it should
"""
file_name = "{0}/{1}/dna".format(self.ensembldb, species.lower())
# get the template name (dependent on the assembly)
tmp_file = ""
for f in os.listdir(file_name):
# check if we've stumbled on the protein file
m = re.findall(".dna", f)
if not m:
continue
tmp_file = f
m = re.findall("(.*).dna", tmp_file)
if masked != 0:
file_name = "%s/%s.dna_rm." % (file_name, m[0])
else:
file_name = "%s/%s.dna." % (file_name, m[0])
if sequence_type == "chromosome":
file_name = "%schromosome.%s.fa" % (file_name, sequence_id)
else:
file_name = "%stoplevel.fa" % (file_name)
if not os.path.exists(file_name):
# find *dna.toplevel.fa file
path = "{0}/{1}/dna".format(self.ensembldb, species.lower())
cmd = "ls " + path + "/*dna.toplevel.fa"
retval = commands.getoutput(cmd)
if retval:
file_name = retval.rstrip()
# otherwise just let the thing fail, I have no better idea
return file_name
def _generate_proteindb_file_name(self, species, protein_type):
"""
@param species: species name (ensembl)
@param protein_type: all / abinitio
@return: protein database name
"""
file_name = "%s/%s/pep" % (self.ensembldb, species.lower())
tmp_file = ""
for f in os.listdir(file_name):
# check if we've stumbled on the protein file
m = re.findall(".pep", f)
if not m:
continue
tmp_file = f
m = re.findall("(.*).pep", tmp_file)
if protein_type == "all":
file_name = "%s/%s.pep.all.fa" % (file_name, m[0])
else:
file_name = "%s/%s.pep.abinitio.fa" % (file_name, m[0])
return file_name