Polymer simulations of compartments and extrusion for modeling the mitosis-to-G1 transition and, more generally, chromatin compartmentalization with or without loop extrusion. See Refs [1-2].
Also see: The polychrom library (https://github.com/open2c/polychrom/) [3], a wrapper for the OpenMM MD package [4]. Beyond using the polychrom library, the code in this repository used examples and methods found in polychrom as a starting point.
There are two simulation codes here:
- comp_extr - This code is used for equilibrated polymer sims (used in parameter sweeps)
- m-to-g1 - A code for performing time-calibrated polymer simulations that progress from mitotic-like chromosomes to interphase-like chromosomes.
Additionally, the examples directory contains data from short, small runs as examples, along with the commands (with command line options) used to run the simulations.
To run steady-state simulations simulations in the compartment/microcompartment configuration used to model the Dag1 locus, provide the configuration files as inputs in the command line. e.g.:
python compSim.py comppath=dag1/comps.dat microcomppath=dag1/microcompsAnaTelo.dat
Other parameters can be set/changed by command line inputs as well. For more details:
python compSim.py ?
M-to-G1 simulations of the Dag1 locus are run similarly (with the addition of an input for CTCF positions):
python m-to-g1_transition.py comppath=dag1/comps.dat microcomppath=dag1/microcompsAnaTelo.dat ctcfpath=dag1/ctcf.dat
Code will run from directory without additional setup (provided that polychrom and OpenMM are also installed). Simulations were originally run on GPUs on machines running Ubuntu 22.04.5 OS.
[1] VY Goel et al. Dynamics of microcompartment formation at the mitosis-to-G1 transition. Nat. Struct. Mol Biol. 32:2614 (2025). https://www.biorxiv.org/content/10.1101/2024.09.16.611917
[2] G Wutz, IF Davidson, EJ Banigan et al. PDS5 proteins control genome architecture by limiting the lifetime of cohesin-NIPBL complexes. Mol. Cell 86:1614 (2026). https://www.biorxiv.org/content/10.1101/2025.08.30.67324
[3] M Imakaev, A Goloborodoko, HB Brandao. polychrom v0.1.0. Zenodo: https://zenodo.org/records/3579473 DOI: 10.5281/zenodo.3579472
[4] P Eastman et al. OpenMM 8: Molecular Dynamics Simulation with Machine Learning Potentials. J Phys Chem B 128:109-116 (2023). https://arxiv.org/abs/2310.03121