def get_positions_to_skip_from_alignment( alignment, fraglen ):
'''
eg. if fraglen equals to 3
sequence FDASFDSA-----FDSAFDSAFDSA
pos_to_acquire ^^^^^^ ^^^^^^^^^^
'''
chainbreak_resnums = []
seq_map_dict = alignment_util.seq_mapping( alignment )
res_nums = sorted( seq_map_dict.keys() )
for idx, rsn in enumerate( res_nums ):
try:
next_rsn = res_nums[idx+1]
#print idx
except IndexError:
# meaning it has reached the end of the container, get termini!
chainbreak_resnums.append( rsn )
ref_rsn = seq_map_dict[rsn]
ref_next_rsn = seq_map_dict[next_rsn]
if ref_next_rsn - ref_rsn > 1:
chainbreak_resnums.append( rsn )
rsd_to_skip = []
for rsd in chainbreak_resnums:
rsd_to_skip += range(rsd-fraglen+2, rsd+1 )
rsd_to_skip = list( set(rsd_to_skip) )
return rsd_to_skip
def align_and_renumber_pdb(fulllength_fasta, truncated_pdbfile, ignore_check=False):
""" This function is going to make the renumber_pdb() obsolete """
# make alignment
fl_seq = seq_util.fasta_file_reader(fulllength_fasta)
tc_seq = seq_util.pdb2fasta(truncated_pdbfile)
alignment = alignment_util.align_two_seqs(fl_seq, tc_seq)
if ignore_check:
pdb_idx1(
truncated_pdbfile, "temp.pdb"
) # for the following step, this has been used in alignment_util.correct_alignment_using_pdb
else:
alignment = alignment_util.correct_alignment_using_pdb(alignment, truncated_pdbfile, False)
seq_map = alignment_util.seq_mapping(alignment)
xyz_dict, pdbline_dict, resname_dict = create_xyzDict_bychain("temp.pdb")
assert len(pdbline_dict.keys()) == 1, (
"this script does not deal with pdbs containing multiple chains (%s)" % pdbline_dict.keys()
)
chain = pdbline_dict.keys()[0]
xyz_dict = xyz_dict[chain]
pdbline_dict = pdbline_dict[chain]
resname_dict = resname_dict[chain]
res_nums = sorted(pdbline_dict.keys())
out_pdblines = "REMARK full_length_aln %s\n" % alignment[0]
out_pdblines += "REMARK truncated_aln %s\n" % alignment[1]
for idx, rsn in enumerate(res_nums):
newrsn = seq_map[rsn]
for line in pdbline_dict[rsn].split("\n")[:-1]: # [:-1], because the last item in the list is ''
out_pdblines += line[0:22] + "%4s" % newrsn + line[26:] + "\n"
out_pdblines += "TER\n"
os.remove("temp.pdb")
return out_pdblines
fl_seq = seq_util.fasta_file_reader(opts.fragfile_fasta)
if opts.truncated_pdb:
tc_seq = seq_util.pdb2fasta(opts.truncated_pdb)
alignment = alignment_util.correct_alignment_using_pdb(
alignment_util.align_two_seqs(fl_seq, tc_seq), opts.truncated_pdb
)
elif opts.truncated_fasta:
tc_seq = seq_util.fasta_file_reader(opts.truncated_fasta)
alignment = alignment_util.align_two_seqs(fl_seq, tc_seq)
else:
sys.stderr.write("ERROR: you need to either give --truncated_pdb or --truncated_fasta\n")
exit()
chainbreak_resnums = frag_util.get_positions_to_skip_from_alignment(alignment, frag_len)
seq_map = alignment_util.seq_mapping(alignment)
residues = sorted(seq_map.keys())
# residues before chain break shouldn't take
# get residue positions to keep
out_fraglines = ""
if opts.debug:
out_fraglines += "# %s\n" % (alignment[0])
out_fraglines += "# %s\n" % (alignment[1])
for pos in residues:
# skip positions at chainbreaks and termini
if pos in chainbreak_resnums:
sys.stderr.write("Skipping: %s\n" % pos)
continue