schrodinger.protein.protassign.network_solver module

This module contains NetworkSolver, a ProtAssign wrapper around toulbar2 module.

Copyright Schrodinger, LLC. All rights reserved.

class schrodinger.protein.protassign.network_solver.NetworkSolver(cluster, interact, upper_bound)

Bases: object

Wrapper around toulbar2 that exactly solves the hydrogen bond network problem

PRECISION = 5
MAX_TIME = 30
TOULBAR2 = 'toulbar2'
__init__(cluster, interact, upper_bound)
Parameters

  • cluster (ProtAssign.hbond_cluster) – Hydrogen bond network cluster that will be optimized

  • interact (Dict[int, Dict[int, bool]]) – Changeable interaction lookup table.

  • upper_bound (float) – Upper energy bound for the network energy. Required for toulbar2

setup_toulbar2_inputs()

Setup the input file for toulbar2. Essentially the file contains data on the variables involved, and the self and pair scores, all in a json file.

parse_and_delete_output_file(out)

Parse and then delete toulbar2 output file that contains the chosen state for each changeable

Parameters

out (str) – File name of output file

Returns

List of chosen state of each changeable

Return type

List[List[int]]

optimal_solution()

Run toulbar2 to get the optimal solution

Returns

Optimal state combination. If toulbar2 fails returns None

Return type

List[int] or None

explore_solutions(upper_bound=None, number=1000)

Run toulbar2 to greedily obtain a number of solutions with an energy below a certain upper bound. Note that there are no guarantees here about diversity, though each solution will be unique.

Parameters

  • upper_bound (float) – upper bound energy. All solutions will have an energy lower than this

  • number (int) – Max number of solutions to find

Returns

List of state combinations

Return type

List[List[int]]