def exercise_validate_sequence():
from iotbx.pdb.amino_acid_codes import validate_sequence
test_sequence = 'abcdefghijklmnopqrstuvwxyz'
a = validate_sequence(test_sequence,
protein=True,
strict_protein=True,
nucleic_acid=False)
assert (len(a) == 3)
a = validate_sequence(test_sequence,
protein=True,
strict_protein=False,
nucleic_acid=False)
assert (len(a) == 1)
a = validate_sequence(test_sequence,
nucleic_acid=True,
strict_nucleic_acid=True,
protein=False)
assert (len(a) == 21)
a = validate_sequence(test_sequence,
nucleic_acid=True,
strict_nucleic_acid=False,
protein=False)
assert (len(a) == 10)
a = validate_sequence(test_sequence,
protein=True,
strict_protein=True,
nucleic_acid=True,
strict_nucleic_acid=True)
assert (len(a) == 3)
def exercise_validate_sequence():
from iotbx.pdb.amino_acid_codes import validate_sequence
test_sequence = "abcdefghijklmnopqrstuvwxyz"
a = validate_sequence(test_sequence, protein=True, strict_protein=True, nucleic_acid=False)
assert len(a) == 6
a = validate_sequence(test_sequence, protein=True, strict_protein=False, nucleic_acid=False)
assert len(a) == 2
a = validate_sequence(test_sequence, nucleic_acid=True, strict_nucleic_acid=True, protein=False)
assert len(a) == 21
a = validate_sequence(test_sequence, nucleic_acid=True, strict_nucleic_acid=False, protein=False)
assert len(a) == 10
a = validate_sequence(test_sequence, protein=True, strict_protein=True, nucleic_acid=True, strict_nucleic_acid=True)
assert len(a) == 5
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
)