def align_chain_sequence(sys_type):
if g_var.args.v >= 2:
print(gen.print_sequnce_info('PROTEIN'))
at = {}
test_chain = {}
for chain_at in range(len(g_var.atomistic_protein_input_raw)):
skip_sequence = False
chain_cg = 0
s = difflib.SequenceMatcher(None,
g_var.seq_at[sys_type][chain_at],
g_var.seq_cg[sys_type][chain_cg],
autojunk=False)
seq_info = s.get_matching_blocks()
while seq_info[0][2] != len(g_var.seq_at[sys_type][chain_at]):
if chain_cg >= len(g_var.seq_cg[sys_type]) - 1:
print(
'\nCannot find a match for user supplied chain: ' +
str(chain_at)
) #+'\n\nAtomistic chain:\n'+str(seq_user[chain_at]),'\n\nIn CG:\n'+str(sequence))
skip_sequence = True
break
if not skip_sequence:
chain_cg += 1
s = difflib.SequenceMatcher(None,
g_var.seq_at[sys_type][chain_at],
g_var.seq_cg[sys_type][chain_cg],
autojunk=False)
seq_info = s.get_matching_blocks()
if not skip_sequence:
temp = {}
if chain_cg not in at:
at[chain_cg] = {}
if seq_info[0][2] == len(g_var.seq_at[sys_type][chain_at]):
for resid, residue in enumerate(
g_var.atomistic_protein_input_raw[chain_at]):
temp[resid +
seq_info[0][1]] = g_var.atomistic_protein_input_raw[
chain_at][residue]
at[chain_cg][str(seq_info[0][1]) + ':' +
str(seq_info[0][1] + seq_info[0][2])] = temp
g_var.seq_cg[sys_type][chain_cg] = mask_sequence(
g_var.seq_cg[sys_type][chain_cg], seq_info[0][1],
seq_info[0][1] + seq_info[0][2])
test_chain[chain_at] = chain_cg
check_chain_alignment_coverage(at, sys_type)
if len(g_var.atomistic_protein_input_raw) < len(g_var.seq_cg[sys_type]):
print(
'### WARNING you have supplied fewer chains than exist in the CG system ###\n'
)
if len(at) > 0:
g_var.user_at_input = True
g_var.atomistic_protein_input_aligned = at
else:
g_var.user_at_input = False
if g_var.group_chains == 'chain':
g_var.group_chains = test_chain
g_var.args.box)
else:
g_var.box_vec = box_vec_initial
box_shift = np.array([0, 0, 0])
read_in.real_box_vectors(g_var.box_vec)
#### pbc fix and residue truncation if required
read_in.fix_pbc(box_vec_initial, g_var.box_vec, box_shift)
#### checks if fragment database and input files match
at_mod.sanity_check()
### convert protein to atomistic representation
g_var.tc['r_i_t'] = time.time()
if 'PROTEIN' in g_var.cg_residues:
g_var.coord_atomistic = at_mod_p.build_multi_residue_atomistic_system(
g_var.cg_residues, 'PROTEIN') ## converts protein to atomistic
if not g_var.user_at_input and g_var.args.v >= 1: ## prints protein sequences
print(gen.print_sequnce_info('PROTEIN'))
## reads in user chain, runs a sequence alignment and finds existing disulphide bonds
g_var.tc['p_d_n_t'] = time.time()
if g_var.user_at_input:
for file_num, file_name in enumerate(g_var.args.a):
atomistic_protein_input_raw, g_var.chain_count = read_in.read_in_atomistic(
g_var.input_directory + 'AT_INPUT_' + str(file_num) +
'.pdb') ## reads in user structure
g_var.atomistic_protein_input_raw.update(
atomistic_protein_input_raw)
read_in.duplicate_chain() ## duplicates user chcains
at_mod_p.check_sequence() ## gets user sequence
at_mod_p.align_chain_sequence('PROTEIN') ## aligns chains
at_mod_p.find_disulphide_bonds_user_sup(
) ## finds user disulphide bonds
at_mod_p.find_disulphide_bonds_de_novo() ## finds CG disulphide bonds