Project ID: plumID:22.032
Name: Reciprocal barrier restraint. Application to path-meta-eABF
Archive: https://zenodo.org/record/7030287/files/CRL-VHL-degrader-SMARCA2-ring-closure-path-meta-eabf.zip
Category: methods
Keywords: restraint, upper wall, lower wall, path colvar, meta-eABF, metadynamics, DRR, protein conformational transition, PROTAC
PLUMED version: 2.7
Contributor: Istvan Kolossvary
Submitted on: 29 Aug 2022
Publication: I. Kolossváry, W. Sherman, Comprehensive Approach to Simulating Large Scale Conformational Changes in Biological Systems Utilizing a Path Collective Variable and New Barrier Restraint. The Journal of Physical Chemistry B. 127, 5214–5229 (2023)
PLUMED input files
| File | Compatible with |
|---|---|
| plumed.dat |   |
Last tested: 22 Apr 2024, 20:44:20
Project description and instructions
To run this simulation you need OpenMM 7.6 or newer and the OpenMM-PLUMED plugin installed. To start the job run the following command line python run_path-meta-eabf.py n_steps do_relax do_plumed. For example, python run_path-meta-eabf.py 50000 1 0 will minimize the system and run 50,000 steps of MD relaxation while python run_path-meta-eabf.py 5000000 0 1 will read the checkpoint file and continue 5 million steps of meta-eABF simulation using the path colvar representing the opening and closing of the full CRL-VHL-degrader-SMARCA2 ring complex. NOTE. The current implementation of the PATHMSD action has a flaw that results in numerical instability due to computing exp(-lambda*(it.distance)) with very large lambda values, without scaling. I highly recommend to make the following simple edit to PathMSDBase.cpp to introduce proper scaling, as follows. $ diff PathMSDBase.cpp PathMSDBase.cpp-orig 275d274 < double min_distance=1e10; 281,285c280 < if(it.distance < min_distance) min_distance=it.distance; < } < < for(auto & it : imgVec) { < it.similarity=exp(-lambda*(it.distance - min_distance)); --- > it.similarity=exp(-lambda*(it.distance)); 293c288 < val_z_path->set(-(1./lambda)*std::log(partition) +min_distance); --- > val_z_path->set(-(1./lambda)*std::log(partition));
Submission history
[v1] 29 Aug 2022: original submission
Badge
Click on the image below and get the code to add the badge to your website!
