Project ID: plumID:24.014
Name: Learning Collective Variables with Synthetic Data Augmentation through Physics-inspired Geodesic Interpolation
Archive: https://github.com/learningmatter-mit/geodesic-interpolation-cv/archive/refs/tags/v0.1.0.zip
Category: methods
Keywords: data augmentation, geodesic interpolation, collective variables, protein folding
PLUMED version: 2.9
Contributor: Juno Nam
Submitted on: 16 Jun 2024
Publication: S. Yang, J. Nam, J. C. B. Dietschreit, R. Gómez-Bombarelli, Learning Collective Variables with Synthetic Data Augmentation through Physics-Inspired Geodesic Interpolation. Journal of Chemical Theory and Computation. 20, 6559–6568 (2024)

PLUMED input files

File Compatible with
…ns/enhanced/plumed_files/plumed_Reg_geo_filter.dat tested on v2.10 tested on master
…/enhanced/plumed_files/plumed_Reg_geo_nofilter.dat tested on v2.10 tested on master
simulations/enhanced/plumed_files/plumed_TDA.dat tested on v2.10 tested on master
…ns/enhanced/plumed_files/plumed_TDA_geo_filter.dat tested on v2.10 tested on master
…/enhanced/plumed_files/plumed_TDA_geo_nofilter.dat tested on v2.10 tested on master
simulations/unbiased/folded/plumed.dat tested on v2.10 tested on master
simulations/unbiased/unfolded/plumed.dat tested on v2.10 tested on master

Last tested: 23 Apr 2025, 10:11:59

Project description and instructions
To deploy the learned CV in MD simulations, you need to build the PLUMED package with the pytorch and drr modules, and then build the GROMACS package with the PLUMED patch. We tested our code with PLUMED 2.9.0 with libtorch 2.0.1 and GROMACS 2023.

Submission history
[v1] 16 Jun 2024: original submission

Badge
Click on the image below and get the code to add the badge to your website!
plumeDnest:24.014