#!/usr/bin/env python3
# -*- coding: utf-8 -*-
''' Align a list of molecules using `super` command in PyMol. The first item
in the list is considered as the reference.
'''
import pymolPy3
import pyrotein as pr
import os
import colorsimple as cs
from loaddata import load_xlsx
# Specify chains to process...
fl_chain = "chains.comp.xlsx"
lines = load_xlsx(fl_chain)
drc = "pdb"
# Define atoms used for distance matrix analysis...
peptide = ["N", "CA", "C", "O"]
# Specify the range of atoms from rhodopsin...
nterm = 1
cterm = 322
len_peptide = (cterm - nterm + 1) * len(peptide)
# Start pymol
pm = pymolPy3.pymolPy3()
## pm("bg white")
# Get the color palette...
color_items = [i[4] for i in lines]
gp("plot \\")
gp(f"'-' using 1:2 with points pointtype 7 linecolor rgb 'blue',\\")
gp("")
for i, v in enumerate(omega.reshape(-1)):
v *= 180 / np.pi
gp(f"{i} {v}")
gp("e")
gp("unset multiplot")
gp("exit")
# [[[ ANALYZE PDB ENTRIES ]]]
# Specify chains to process...
fl_chain = "chains.comp.xlsx"
lines = load_xlsx(fl_chain, sheet="DesKC", splitchain=True)
# Select residues...
nterm = 170
cterm = 206
## cterm = 200
sample_per_rise = 1
num_turn = (cterm - nterm + 1)
# Consider offset...
offterm = 4
nterm_aux = nterm - offterm
cterm_aux = cterm + offterm
num_turn += offterm * 2
# The range to highlight...
# User defined range...
nterm, cterm = 1, 322
len_seg = cterm - nterm + 1
super_seg = super_seq[nseqi:nseqi + len_seg]
# Load constant -- atomlabel...
label_dict = pr.atom.constant_atomlabel()
aa_dict = pr.atom.constant_aminoacid_code()
# Calculate the total length of distance matrix...
len_dmat = np.sum([len(label_dict[aa_dict[i]]) for i in super_seg])
# [[[ ANALYZE PDB ENTRIES ]]]
# Specify chains to process...
fl_chain = "chains.comp.xlsx"
lines = load_xlsx(fl_chain, sheet="Sheet1")
drc = "pdb"
## dmats = np.zeros((len(lines), len_dmat, len_dmat))
len_lower_tri = (len_dmat * len_dmat - len_dmat) // 2
dmats = np.zeros((len(lines), len_lower_tri))
# Process each entry...
for i_fl, line in enumerate(lines[:]):
# Unpack parameters
_, pdb, chain, species = line[:4]
print(f"Processing {pdb}_{chain}")
# Read coordinates from a PDB file...
fl_pdb = f"{pdb}.pdb"
pdb_path = os.path.join(drc, fl_pdb)
def reverse_sign(u, vh, rank, index_from_zero = True):
# Comply with the convention (1-based index)
rank_in_data = rank if index_from_zero else rank - 1
# Reverse sign...
u[:, rank_in_data] = - u[:, rank_in_data]
vh[rank_in_data, :] = -vh[rank_in_data, :]
return None
# Specify chains to process...
fl_chain = "rhodopsin.db.xlsx"
lines = load_xlsx(fl_chain, sheet = "total", splitchain = True)
# Load upstream data...
u = np.load("u.seq.npy")
s = np.load("s.seq.npy")
vh = np.load("vh.seq.npy")
len_seq = np.load("len_seq.npy")
# Allow positive value that fits the distance in nature (optinal)...
reverse_sign(u, vh, 1, index_from_zero = False)
## reverse_sign(u, vh, 2, index_from_zero = False)
## reverse_sign(u, vh, 4, index_from_zero = False)
## reverse_sign(u, vh, 6, index_from_zero = False)
# Calculate the weighted coefficients...
# The following is a computational equivalence to c = RMS(u @ s, axis = 0) @ vh
