def run (args=(), params=None, out=sys.stdout) :
assert (params is not None)
seq_files = params.muscle.seq_file
output_file = params.muscle.output_file
if (output_file is None) or (output_file == "") :
output_file = os.path.join(os.getcwd(), "muscle.aln")
from iotbx import file_reader
from iotbx.bioinformatics import any_sequence_format, sequence
seqs = []
for file_name in seq_files :
if (file_name.endswith(".pdb") or file_name.endswith(".ent") or
file_name.endswith(".pdb.gz") or file_name.endswith(".ent.gz")) :
pdb_in = file_reader.any_file(file_name, force_type="pdb").file_object
hierarchy = pdb_in.hierarchy
first_model = hierarchy.models()[0]
found_protein = False
for chain in first_model.chains() :
if chain.is_protein() :
chain_seq = chain.as_padded_sequence()
base_name = os.path.basename(file_name)
seq_name = "%s_%s" % (os.path.splitext(base_name)[0], chain.id)
seqs.append(sequence(chain_seq, seq_name))
found_protein = True
if (not found_protein) :
raise Sorry(("The PDB file %s does not contain any recognizable "+
"protein chains.") % file_name)
else :
try :
seq_objects, non_compliant = any_sequence_format(file_name,
assign_name_if_not_defined=True)
seqs.extend(seq_objects)
except Exception, e :
raise Sorry(("Error parsing '%s' - not a recognizable sequence "+
"format. (Original message: %s)") % (file_name, str(e)))
def run (args=(), params=None, out=sys.stdout) :
assert (params is not None)
seq_files = params.muscle.seq_file
output_file = params.muscle.output_file
if (output_file is None) or (output_file == "") :
output_file = os.path.join(os.getcwd(), "muscle.aln")
from iotbx import file_reader
from iotbx.bioinformatics import any_sequence_format, sequence
seqs = []
for file_name in seq_files :
if (file_name.endswith(".pdb") or file_name.endswith(".ent") or
file_name.endswith(".pdb.gz") or file_name.endswith(".ent.gz")) :
pdb_in = file_reader.any_file(file_name, force_type="pdb").file_object
hierarchy = pdb_in.hierarchy
first_model = hierarchy.models()[0]
found_protein = False
for chain in first_model.chains() :
if chain.is_protein() :
chain_seq = chain.as_padded_sequence()
base_name = os.path.basename(file_name)
seq_name = "%s_%s" % (os.path.splitext(base_name)[0], chain.id)
seqs.append(sequence(chain_seq, seq_name))
found_protein = True
if (not found_protein) :
raise Sorry(("The PDB file %s does not contain any recognizable "+
"protein chains.") % file_name)
else :
try :
seq_objects, non_compliant = any_sequence_format(file_name,
assign_name_if_not_defined=True)
seqs.extend(seq_objects)
except Exception, e :
raise Sorry(("Error parsing '%s' - not a recognizable sequence "+
"format. (Original message: %s)") % (file_name, str(e)))
def get_muscle_alignment_ordered(sequences, out = None):
from iotbx import bioinformatics
name_for = {}
for ( i, seq ) in enumerate( sequences, start = 1 ):
name = name_for.get( seq, "Chain_%d" % i )
name_for[ seq ] = name
alignment = get_muscle_alignment(
fasta_sequences = "\n".join(
str( bioinformatics.sequence( name = name, sequence = seq.sequence ) )
for ( seq, name ) in name_for.items()
),
out = out,
)
lookup = dict( zip( alignment.names, alignment.alignments ) )
assert all( n in lookup for n in name_for.values() )
return bioinformatics.clustal_alignment(
names = [ seq.name for seq in sequences ],
alignments = [ lookup[ name_for[ seq ] ] for seq in sequences ],
program = alignment.program
)
def get_muscle_alignment_ordered(sequences, out = None):
from iotbx import bioinformatics
from iotbx.pdb.amino_acid_codes import validate_sequence
name_for = {}
for ( i, seq ) in enumerate( sequences, start = 1 ):
name = name_for.get( seq, "Chain_%d" % i )
name_for[ seq ] = name
alignment, errors = get_muscle_alignment(
fasta_sequences = "\n".join(
str( bioinformatics.sequence( name = name, sequence = seq.sequence ) )
for ( seq, name ) in name_for.items()
),
out = out,
)
# check for errors and handle:
# invalid characters in sequences
if (len(errors) > 0):
for error in errors:
error = error.strip()
if ('Invalid character' in error):
for seq in name_for.keys():
invalid = validate_sequence(
seq.sequence, protein=True, strict_protein=False,
nucleic_acid=True, strict_nucleic_acid=False)
if (len(invalid) > 0):
name_for.pop(seq)
sequences = name_for.keys()
elif (len(error) > 0):
raise Sorry(error)
lookup = dict( zip( alignment.names, alignment.alignments ) )
assert all( n in lookup for n in name_for.values() )
return bioinformatics.clustal_alignment(
names = [ seq.name for seq in sequences ],
alignments = [ lookup[ name_for[ seq ] ] for seq in sequences ],
program = alignment.program
)