Rare-Event-Sampling-Gene-Networks

This repository contains code for running the weighted ensemble (WE) sampling method with adaptive Voronoi-based discretization on biochemical networks.

Requirements

• Operating System Requirements/Recommendations
  • Linux (Recommended) or Mac OS X
  • Windows is not fully implemented but a basic version is included as an example
• Matlab 2013a or newer

There are three versions of the code, one for running the WE method to find a transition matrix of transitions between Voronoi-based sampling regions (General_bin_movement_and_transition_local_linux.m and General_bin_movement_and_transition_local_windows.m), and two for running the WE method to find the transition rate between two states of interest.

Included

There are three versions of the code (found in the Linux folder):

1. Running the WE method to find a transition matrix of transitions between Voronoi-based sampling regions
   • General_bin_movement_and_transition_local.m
2. Running the WE method to find the transition rate between two hypershperes.
   • General_bin_movement_and_rates_hypersphere_local.m
3. Running the WE method to find the transition rate between two sets of Voronoi centers.
   • General_bin_movement_and_rates_voronoi_bins_local.m

Instructions

Permissions to read the BioNetGen run_network application (under /BioNetGen_files/) and any *.net files must be properly set.

To run the code, use submission_script.sh or submission_script.m for linux or windows, respectively.

Inside submission_script.sh (or submission_script.m), all parameters for running the WE method can be set.

Comment/uncomment the appropriate lines to select which version (1-3) to run. The default is the transition matrix calculation.

The output of General_bin_movement_and_transition_local_linux.m is a .mat file with the prefix specified in the input parameters (see submission_script.m or submission_script.sh).

Inside the .mat file are the following variables:

• VoronoiCenters: the current position of all Voronoi centers

• replicas: the current replica positions, tags, and replica weights

• transition_matrix_count: the transition matrix calculated by finding the counts of transitions from sampling region i to sampling region during every lagtime tau

• transition_matrix_count: the transition matrix calculated by summing the weight of transitions from sampling region i to sampling region during every lagtime tau

• iteration: the current 0-4 temporary iteration number for temporarily stored replica information

• toc: the current total computational time

• ijk: the current ijk iteration of transition matrix calculation