evanfeinberg

def fix_topology(topology):
	
	new_top = topology.copy()

	residues = {}
	for chain in new_top.chains:
		#print chain
		for residue in chain.residues:
			resname = str(residue)
			if resname in list(residues.keys()):

evanfeinberg

from sklearn.ensemble import RandomForestClassifier, RandomForestRegressor
import numpy as np 

def compute_random_forests(features, projected_tica_coords, rf_dir, n_trees=500, 
  n_tica_components=25, start_tIC=0):

  features = np.concatenate(features)
  tics = np.concatenate(projected_tica_coords)

  for j in range(start_tIC, np.shape(tics)[1]):

evanfeinberg

"""
The :mod:`sklearn.kernel_approximation` module implements several
approximate kernel feature maps base on Fourier transforms.
"""

# Author: Andreas Mueller <amueller@ais.uni-bonn.de>
#
# License: BSD 3 clause

import warnings

evanfeinberg

def fix_topology(topology):
	new_top = topology.copy()

	residues = {}
	for chain in new_top.chains:
		#print chain
		for residue in chain.residues:
			resname = str(residue)
			if resname in residues.keys():
				residues[resname].append(residue)

evanfeinberg

import mdtraj as md
import os
import h5py
import multiprocessing as mp
from functools import partial 
import pytraj.io as mdio

def get_trajectory_files(traj_dir, ext = ".h5"):
	traj_files = []
	for traj in os.listdir(traj_dir):

evanfeinberg

def fix_topology(topology):
	
	new_top = topology.copy()

	residues = {}
	for chain in new_top.chains:
		#print chain
		for residue in chain.residues:
			resname = str(residue)
			if resname in residues.keys():

evanfeinberg

'''
Many PDB Files have an issue where a single chemical residue (e.g., a single histidine) will have each of its atoms, while
	chemically bonded, spread out across the PDB file. This confuses many PDB readers, as it will assume each non-contiguous
	line for that residue corresponds to separate residues. This class has a function that takes as inputs the name of the original
	pdb file and the name of a new file to which the user would like to write. It then finds all residues corresponding to proteins,
	and stores them in a dict, with keys as a tuple of (residue_name, residue_chain, and residue_number) and with values as a 
	list of physical lines in the PDB file that correspond to that unique residue identifier. The function then prints the initial header,
	prints the exact same PDB file lines just in a different order that maintains contiguousness within each residue, and then prints
	all membrane, solvent, and other lines.

evanfeinberg

'''I should formalize this with a class, etc., but here's how one can use this: 

with a file corresponding to a trajectory you want to featurize, call:

featurize_custom(traj_file)

'''


def fix_topology(topology):