patchyAnalysisTools

A set of tools to analyze the output of patchy particle simulations. Works with a very specific format of trajectory files, but the I/O methods can be extended for the analysis tools to work on arbitrary trajectory file formats. See examples for details.

Installation

git clone https://github.com/sodiumnitrate/patchyAnalysisTools.git
cd patchyAnalysisTools
python setup.py install

Examples

(See the examples folder for more details.)

Loading the module

You can load the module as

import patchyAnalysisTools.trajectory as trj

The following examples all use the alias trj for the module.

Loading trajectories

To load a trajectory:

traj = trj.trajectory(file_name="traj.conf")

where traj.conf is a trajectory file with specific formatting requirements (see examples). Patch information is then read from a file and set for the traj object by

traj.set_patch_info("patches.dat")

where patches.dat is again a special file (see examples). Some analysis can be done without reading and setting patch information, but most require this for determining bonds, etc.

Selecting frames or subtrajectories

To select the last frame:

last_frame = traj.get_last_frame()

To select an arbitrary frame:

frame = traj.get_frame(10)

To select a range of frames:

subtraj = traj.get_frame(10,20)

Bonding and cluster info

Determination of bonding and cluster information requires the patches to have been set in the trajectory before a frame is selected (see above).

First, we determine bonds:

frame.get_list_of_interacting_pairs()

If the trajectory file has the number of bonds per particle written, we can double check the calculated bonds against these values:

frame.check_calculated_bonds_against_bond_numbers()

We then determine the clusters by

frame.set_cluster_info()

To find the percolating clusters:

frame.find_percolating_clusters()

To check if system is percolated:

frame.is_system_percolated()

which returns a boolean.

What's missing

The main thing that is incomplete in this package is support for the output of Gibbs-Ensemble simulations (which contain two boxes). Some of the tools in this package support it, some don't. I may complete it at one point, but for now, everything is only tested for regular simulation outputs.

It is also important to note that not all the tools in this package are thoroughly tested for simulation output with non-cubic boxes.