Examine the relationship between the sidechain separation and protonation

analyze_prontonation.py

#!/usr/bin/env python
# -*- coding: UTF-8 -*-
from marshal import dump, load
from random import sample
from glob import glob
from pdb import *
import sys


def run_experiment(residues, interests, ss_code=None, mind=0.0, maxd=100.0):
  results = []
  for interest in interests:
    left, right = [], []
    res_name, a1n, a2n, a3n, a4n = interest
    matching_residues = filter(lambda r: r['resName'] == res_name, residues)
    for residue in matching_residues:
      if ss_code != None and residue['ssCode'] != ss_code: continue
      distance = residue_distance(residue, a3n, a4n)
      if distance == None:
        # print("Error on %s (%.2f): " % (res_name, residue['resNum']), find_atoms('HD2', residue))
        continue
      if len(find_atoms('HD2', residue)) > 0:
        left.append(distance)
      else: right.append(distance)
    results.append([res_name, left, right])
  return results

def analyze_result(name, data, mind=0.0, maxd=1000.0, nbins=50):
  print("Average: %.2f Angstrom" % average(data))
  print("Median:  %.2f Angstrom" % median(data))
  rdata = filter(lambda x: mind < x < maxd, data)
  figure(figsize=[11, 8.5])
  hist(rdata, histtype='bar', bins=nbins, label=name)
  legend()
  grid()
  savefig(name + '.png', dpi=300)

interests = [
  ['ASP', 'CB', 'CG', 'OD1', 'OD2'],
  # ['GLU', 'CG', 'CD', 'OE1', 'OE2'],
  # ['ASN', 'CB', 'CG', 'OD1', 'ND2'],
  # ['GLN', 'CG', 'CD', 'OE1', 'NE2'],
  # ['VAL', 'CA', 'CB', 'CG1', 'CG2']
]

try:
  __IPYTHON__
  is_ipy = True
except: is_ipy = False

all_residues = []
for pdb_fname in glob('nrpdb_hd2/*'):
  all_residues += extract_models(pdb_fname)[0]['residues']
r = run_experiment(all_residues, interests)[0]
print("Protonated: %d, Non-protonated: %d" % (len(r[1]), len(r[2])))