Molecular Motor Fibre Toolkit

CONTACT: mattias.akke@gmail.com

Living muscles amplify nanometre-scale actin–myosin motion into centimetre-scale work. Artificial molecular muscles based on Feringa-type rotary motors follow the same principle: amphiphilic motors self-assemble into worm-like nanofibres, bundle into hydrated strings, and bend toward UV light when millions of motors rotate in concert. Small chemical tweaks, however, can disrupt this hierarchy, and connecting molecular design to macroscopic actuation still demands a multiscale modelling pipeline. This repository delivers that pipeline: starting from all-atom force-field derivation, through fibre assembly, coarse-grained mapping, bundle simulations with synthetic rotations, and finally MDAnalysis- based characterisation.

Each experimental phase lives in its own subdirectory with a README documenting local tooling and automation scripts. Combine them to trace how a molecular motor’s chemistry propagates to fibre-scale mechanics and bundle response.

Environments and Requirements

• environment.yml – Conda environment used for analysis notebooks and MDAnalysis-based scripts.
• gromacs_info.txt / plumed_info.txt – versions used for every production simulation (GROMACS 2021-modified with CUDA 11.6, PLUMED 2.7.1 built with MPI). The MDAnalysis tooling was compiled against PLUMED 2.9.2 and GROMACS 2024.5 for post-processing; reproduce those builds if you need to re-run the notebooks.

Project Layout

Directory	Description	README
mkforce_field/	Amber/ParmEd wrappers that derive AA force fields for new motor variants.	mkforce_field/README
mkfiber/	Builds single fibres from AA motors, compresses, solvates, equilibrates.	mkfiber/README
characterise_fibre/	Cleans mkfiber trajectories and runs MDAnalysis/Voronota analysis.	characterise_fibre/README
mkfiber_cluster/	Packmol + CG bundle builder for 5-layer fibre packs and diffusion studies.	mkfiber_cluster/README
bend_fibre/	Generates shifted/rotated 3-fibre bundles to study bent motors.	bend_fibre/README
coarse_grain_fibre/	AA vs CG fibre comparison, SASA, diffusion of bundled systems.	coarse_grain_fibre/README
coarse_grain_MM2/	Full AA→CG pipeline for the MM2 motor core (mapping, fitting, SASA).	coarse_grain_MM2/README
coarse_grain_FMM/	Functional-group analogue of the MM2 pipeline (MM2a motors).	coarse_grain_FMM/README
dimerisation PMF/	PMF calculations for AA & CG motor dimers.	(see folder notes)
annealing/	Annealing protocol tests for AA fibres.	(experimental)
structure_files/	Shared AA/CG structures, topologies, mappings (see below).	—

Each README documents the local run_all scripts, inputs, outputs, and known issues. Follow the sequence mkforce_field → mkfiber → characterise_fibre before branching into the coarse-graining or bundle studies.

Top-Level run_all.sh

1. mkforce_field/bin/run_molecule.sh  # regenerates AA force fields
2. mkfiber/bin/run_all.sh             # builds and equilibrates fibres
3. characterise_fibre/bin/run_all.sh  # cleans & analyses the trajectories

Set MOLTYPE and GPU_ID inside run_all.sh to control which motor subtype is built and which GPU to target. Each internal script has additional flags for fine-grained control if you need to deviate from the default pipeline.

structure_files/ Overview

• gro/, pdb/, top/ – canonical AA structures and topologies for the motor fragments, used by mkforce_field and mkfiber.
• mol2/ – source fragments for AmberTools (head/core/legs) in various subtypes.
• number_elements_by_residue.py, sort_mol2_bonds.pl – helper scripts for pre-processing input files.
• Mapping/CG assets (e.g., mapping_MM2*.ndx, CG solvent boxes) live alongside the directories listed above and are consumed by the coarse-graining modules.