rust-simulation-tools

name	rust-simulation-tools
description	Overview and entry point for the rust_simulation_tools Python package (high-performance MD analysis written in Rust/PyO3). Use when the user wants to install/build the package, load AMBER topologies/coordinates/trajectories, or needs to know which capability-specific skill (system building, ligand parameterization, trajectory analysis, MM-PBSA, minimization, ipSAE) to reach for.

rust_simulation_tools

rust_simulation_tools (import alias rst) is a high-performance molecular dynamics toolkit written in Rust and exposed to Python via PyO3. It parses AMBER files, runs VMD-style selections, computes SASA/alignment, builds simulation systems, parameterizes ligands, minimizes energy, and computes MM-PBSA/GBSA binding free energies — all without OpenMM, AmberTools, or MDAnalysis at runtime.

Installation

From PyPI:

uv pip install rust-simulation-tools

From source (this repo) — requires maturin and Python dev headers:

uv venv .venv && source .venv/bin/activate
uv pip install maturin numpy pytest
maturin develop --release          # builds crates/bindings and installs into the venv

After editing any Rust source, re-run maturin develop --release to rebuild. Run the test suite with pytest tests/ -v.

Importing

import rust_simulation_tools as rst
# or pull specific names
from rust_simulation_tools import read_prmtop, read_inpcrd, DcdReader

Loading a system (the common starting point)

topo = rst.read_prmtop("system.prmtop")     # -> AmberTopology
coords, box = rst.read_inpcrd("system.inpcrd")  # coords: (n_atoms, 3) ndarray
print(topo.n_atoms, topo.n_residues)

dcd = rst.DcdReader("trajectory.dcd")
trajectory, boxes = dcd.read_all()           # read everything into memory
trajectory = trajectory.reshape(dcd.n_frames, dcd.n_atoms, 3)
# AMBER ASCII trajectories: rst.MdcrdReader(path, n_atoms, has_box)

AmberTopology exposes .n_atoms, .n_residues, .atom_names, .residue_labels, .charges(), .sigmas(), .epsilons(), .bonds(), .select(expr, coordinates=None), and .build_selection(residue_indices).

Which skill to use

Task	Skill
Build a solvated/implicit system, add ions, write prmtop/inpcrd	rst-build-system
Assign GAFF2 types + AM1-BCC/Gasteiger charges; raw AM1 charges	rst-parameterize-ligand
Atom selections, SASA, alignment, unwrapping, interaction fingerprints	rst-trajectory-analysis
MM-PBSA / MM-GBSA binding free energy, per-residue decomposition, entropy	rst-mmpbsa
Energy minimization of a prmtop/inpcrd	rst-minimize
ipSAE / pDockQ / LIS interface confidence scoring	rst-ipsae

Runnable example scripts for every workflow live in examples/ at the repo root.