Coarse-graining Approach to Liquid-liquid phase separation Via an Automated Data-driven Optimisation Scheme
This repository contains Python code to run coarse-grained molecular dynamics simulations of intrinsically disordered proteins (IDPs) using the CALVADOS model.
single_chain/
Python code to run single-chain simulations of IDPs using the CALVADOS model.python submit_local.py
runs a simulation of a single ACTR chain on a single CPU.direct_coexistence/
Python code to run multi-chain simulations of IDPs using the CALVADOS model in slab geometry.python submit.py
submits a direct-coexistence simulation of 100 A1 LCD chains on a single GPU.
In the examples, direct-coexistence and single-chain simulations are performed using openMM and HOOMD-blue installed with mphowardlab/azplugins, respectively.
To run the code, install Miniconda and make sure all required packages are installed by issuing the following terminal commands
conda env create -f environment.yml
source activate calvados
Commands to install HOOMD-blue v2.9.3 with mphowardlab/azplugins v0.11.0
curl -LO https://github.com/glotzerlab/hoomd-blue/releases/download/v2.9.3/hoomd-v2.9.3.tar.gz
tar xvfz hoomd-v2.9.3.tar.gz
git clone https://github.com/mphowardlab/azplugins.git
cd azplugins
git checkout tags/v0.11.0
cd ..
cd hoomd-v2.9.3
mkdir build
cd build
cmake ../ -DCMAKE_INSTALL_PREFIX=<path to python> \
-DENABLE_CUDA=ON -DENABLE_MPI=ON -DSINGLE_PRECISION=ON -DENABLE_TBB=OFF \
-DCMAKE_CXX_COMPILER=<path to g++> -DCMAKE_C_COMPILER=<path to gcc>
make -j4
cd ../hoomd
ln -s ../../azplugins/azplugins azplugins
cd ../build && make install -j4
Thea K. Schulze (@theaschulze)
Kresten Lindorff-Larsen (@lindorff-larsen)
G. Tesei, T. K. Schulze, R. Crehuet, and K. Lindorff-Larsen. PNAS 118(44), 2021. DOI: 10.1073/pnas.2111696118
G. Tesei and K. Lindorff-Larsen. Open Research Europe 2023 2(94). DOI: 10.12688/openreseurope.14967.2