FishMol

A pure-Python package for post-processing molecular dynamics trajectories.

                  Welcome!                    ▄▄█▀                  FishMol
                                          ▄▄███▀                 version 0.0.1
      ○                                ▄▄█████▀                          ○
           ○                        ▄▄████████▄                         /
                                ▄▄▄████████████▄                    ○--○
         ○                ▄▄▄████████████████████▄▄                     \ 
                    ▄▄▄██████▀███████████████████████▄▄                  ○--○           ▄
                ▄▄████████████ ██████████████████████████▄▄             /           ▄▄█▀
         ○   ▄█████████████████ ████████████████████████████▄▄         ○         ▄███▀
          ▄██████████▀▀▀████████ ██████████████████████████████▄▄             ▄█████▀
         ▄████████▀   ○  ▀███████ █████████████████████████████████▄▄      ▄▄█████▀
         ■▄███████▄      ▄███████ █████████████████████████████████████▄▄▄███████▀
          ▀█████████▄▄▄█████████ ███████████████████████████████████████████████
            ▀██████████████████ ████████████████████████████████████████████████▄
              ▀██████████████▀▄███████████████████████████████████▀▀▀▀▀████████████▄
                ▀▀█████████▀▄███████████████████████████████▀▀           ▀▀████████▀
                   ▀▀█████▄████████████████████████▀▀▀▀██▀                 ▀▀████▀
                        ▀▀▀█████████████████▀▀▀▀        ▀■                    ▀█▀
                             ▀▀▀███████▀▀
                                 ▀████
                                    ▀▀▄                Contact: Lei.Lei@durham.ac.uk

Features

• Trajectory I/O — fast memory-mapped reading of extended XYZ files; write filtered trajectories back to disk
• PBC-aware geometry — distances, angles, dihedrals, and vectors under the minimum image convention for triclinic cells
• Molecule recognition — automatic clustering of atoms into molecules via a covalent-radius bonding graph
• Distribution functions — radial (RDF), angular (ADF), dihedral (DDF), and combined 2-D (CDF) distribution functions
• MSD & diffusion — FFT-based mean square displacement; self-diffusion coefficient via the Einstein relation
• Diffusion anisotropy — spherical projection of 1-D MSD to map the directional diffusion surface; Voronoi channel assignment
• Hydrogen bonds — frame-by-frame H-bond recognition with distance + angle criteria; biexponential lifetime from autocorrelation
• Vector Reorientation Dynamics — Legendre polynomial autocorrelation (l = 1, 2, 3) with KWW stretched-exponential fitting
• Dimer analysis — dimer lifetime–displacement distribution function (DLDDF)
• Visualisation — interactive 3-D viewer embedded in Jupyter via the ASE backend

Requirements

• Python ≥ 3.8
• NumPy ≥ 1.21
• SciPy ≥ 1.7
• matplotlib ≥ 3.5
• recordclass ≥ 0.17.2
• iteration_utilities ≥ 0.11.0
• colour ≥ 0.1.5
• ASE ≥ 3.22.1 (optional, for 3-D visualisation)

Installation

From PyPI (recommended)

pip install fishmol

To include the optional 3-D visualisation backend (ASE):

pip install fishmol[vis]

Verify the installation:

fishmol

From source

For the latest development version or to contribute:

git clone https://github.com/Lei-Lei-alpha/fishmol.git
cd fishmol
pip install -e ".[vis]"   # omit [vis] if you don't need visualisation
fishmol

Quick Start

from fishmol import trj, funcs

# Define the simulation cell (lattice vectors as rows, in Å)
cell = [
    [21.2944,  0.0000,  0.0000],
    [-4.6030, 20.7909,  0.0000],
    [-0.9719, -1.2106, 15.1054],
]

# Load a trajectory (5 fs timestep, all frames)
traj = trj.Trajectory(timestep=5, data="trajectory.xyz", index=":", cell=cell)

# Calibrate centre-of-mass drift, wrap into box
traj.calib().wrap2box()

# Compute radial distribution function between O and H atoms
o_idx = [i for i, s in enumerate(traj.frames[0].symbs) if s == "O"]
h_idx = [i for i, s in enumerate(traj.frames[0].symbs) if s == "H"]
rdf = funcs.RDF(traj, at_g1=o_idx, at_g2=h_idx, nbins=200, range=(0.0, 8.0))
results = rdf.calculate(plot=True)

Documentation

Full worked examples are in the online documentation: