DiffusiveCohesin

This repository contains the code used for the paper:

"Symmetry and force response of cohesin loop extrusion are determined by diffusion of its motor and anchor domains"

Overview

This repository provides:

• four LAMMPS input scripts corresponding to the four simulation videos in the paper
• one LAMMPS data file for the initial simulation configuration
• custom LAMMPS source files required for the simulations

The simulations were developed for LAMMPS 22Jul2025 on Linux.

Repository contents

Main files in this repository include:

• F0.005_600_604.data
  Initial system configuration

• in.flow_328678_DD14p5_MD14p5.lam

• in.flow_328701_DD14p5_MD14p5.lam

• in.flow_328708_DD14p5_MD14p5.lam

• in.flow_329079_DD14p5_MD24p5.lam
  Example LAMMPS input scripts used for the simulations

• compute_nearest_dna.cpp

• compute_nearest_dna.h

• pair_morse_dynamic_window.cpp

• pair_morse_dynamic_window.h
  Custom LAMMPS source files required by the input scripts

Requirements

• Linux system
• LAMMPS 22Jul2025
• C++ compiler
• make
• optional: MPI if you want to build the MPI version

Installation

1. Download and unpack LAMMPS

Place the LAMMPS 22Jul2025 source archive in your local working directory and unpack it:

tar -xzf lammps-22Jul2025.tar.gz

2. Copy the custom source files into the LAMMPS source directory

Copy all custom .cpp and .h files from this repository into the src directory of LAMMPS:

cp *.cpp *.h /your/local/path/lammps-22Jul2025/src/

Replace /your/local/path/ with your local path.

3. Compile LAMMPS

Go to the LAMMPS source directory:

cd /your/local/path/lammps-22Jul2025/src

Clean old build files:

rm -rf Obj_mpi Obj_serial
rm -f lmp_mpi lmp_serial

Enable the required package:

make yes-molecule

Build the serial executable:

make serial -j8

Optionally, also build the MPI executable:

make mpi -j8

Running a simulation

Open any simulation subdirectory or use any input file in this repository.

For example, to run a simulation on a single CPU core:

/your/local/path/lammps-22Jul2025/src/lmp_serial -in in.flow_328678_DD14p5_MD14p5.lam

To save the output to a log file:

/your/local/path/lammps-22Jul2025/src/lmp_serial -in in.flow_328678_DD14p5_MD14p5.lam | tee run.log

Notes

• Replace all example paths with your local paths.
• The custom pair style and compute must be compiled into LAMMPS before running the input scripts.
• For testing and reproduction, running on 1 CPU core is sufficient.
• Using 1 or 2 CPU cores is recommended for simple runs.

Reproducibility

To reproduce the simulations:

1. install LAMMPS 22Jul2025
2. copy the custom source files into the LAMMPS src folder
3. compile LAMMPS
4. run one of the provided in.*.lam scripts with the provided data file

Contact

For questions about the code or simulations, please contact the repository author.