def Get_Pairwise(m1, m2):
"""Complete a pairwise global alignment of both model's sequence.
Input
-model1, model2
Output:
max_pair = return pair of sequences with highest alignment score
"""
try:
ppb = pdb.CaPPBuilder()
for polypeptide in ppb.build_peptides(m1):
sequence_ref = polypeptide.get_sequence()
for polypeptide in ppb.build_peptides(m2):
sequence_sample = polypeptide.get_sequence()
align = pairwise2.align.globalxx(sequence_ref, sequence_sample)
max_pair = max(align,key = lambda x:x[2])
return max_pair
except UnboundLocalError: #this error raises when the two compared chains are of diferent kind
return None
def bonus_9_2(chain1, chain2, struct1, struct2):
"""
handling the case that the number of atoms is different by align the amino-acids and remove the unnecessary atoms.
:param chain1: the first protein's chain
:param chain2: the second protein's chain
:param struct1: the first protein
:param struct2: the second protein
:return: filtered lists of atoms to align
"""
ppbils = pdb.CaPPBuilder()
peptide1, peptide2 = [], []
filter_peptide_by_chain(chain1, peptide1, ppbils, struct1)
filter_peptide_by_chain(chain2, peptide2, ppbils, struct2)
# converting list to peptide
peptide1 = pdb.Polypeptide.Polypeptide(peptide1)
peptide2 = pdb.Polypeptide.Polypeptide(peptide2)
# converting peptide to sequence and CA list
seq1 = peptide1.get_sequence()
atoms1 = peptide1.get_ca_list()
seq2 = peptide2.get_sequence()
atoms2 = peptide2.get_ca_list()
# align the sequences
alignments = pairwise2.align.globalxx(seq1, seq2)
# filter the atoms lists
ignore_inx1 = [
i for i in range(len(alignments[0][0])) if alignments[0][0][i] == "-"
]
ignore_inx2 = [
i for i in range(len(alignments[0][1])) if alignments[0][1][i] == "-"
]
atoms2 = [atoms2[i] for i in range(len(atoms2)) if i not in ignore_inx1]
atoms1 = [atoms1[i] for i in range(len(atoms1)) if i not in ignore_inx2]
return atoms1, atoms2
def __init__(self, filepath):
"""Constructor of one pdb file : PDBFile.
Arguments :
------------
filepath : string
path to the pdb file
"""
# -----
# save id extracted from path :
self.id = filepath[-8:-4]
# -----
# init parser :
parser = PDB.PDBParser()
struct = parser.get_structure("", filepath)
# -----
# extract from header :
self.keywords = struct.header['keywords']
self.name = struct.header['name']
self.head = struct.header['head']
self.deposition_date = struct.header['deposition_date']
self.release_date = struct.header['release_date']
self.structure_method = struct.header['structure_method']
self.resolution = struct.header['resolution']
self.structure_reference = str(struct.header['structure_reference'])
self.journal_reference = struct.header['journal_reference']
self.author = struct.header['author']
self.compound = str(struct.header['compound'])
# -----
# Get the sequence and the angles
# extract all polypeptides from the structure :
ppb = PDB.CaPPBuilder()
# The sequence of each polypeptide can then easily be obtained
# from the Polypeptide objects :
self.seq = ""
atom_idx = 0
start = 0
end = 0
for pp, chain in zip(ppb.build_peptides(struct), struct.get_chains()):
print (pp)
seq = str(pp.get_sequence())
# The sequence is represented as a Biopython Seq object,
# and its alphabet is defined by a ProteinAlphabet object.
print (seq)
self.seq += seq
# Get the boundary of the peptide
# using residu id
# A residue id is a tuple with three elements:
# - The hetero-flag
# - *The sequence identifier in the chain*
# - The insertion code,
# start of the polypeptide : pp[0].get_id()[1]
# end of the polypeptide : pp[-1].get_id()[1]
start = end + 1
print (start)
end = start + len(seq)-1
print (end)
# |-----------||-------------------|
# sA sA sB eB
self.chains.append(Chain(chain.id, self.id, start, end))
# Get phi psi angle
angles = pp.get_phi_psi_list()
# Some are None because :
# - Some atoms are missing
# -> Phi/Psi cannot be calculated for some residue
# - No phi for residue 0
# - No psi for last residue
print(angles)
for phi, psi in angles:
atom_idx += 1
self.angles.append(Angle(self.id, atom_idx, phi, psi))