PaCS-Toolkit User Guide

Welcome to PaCS-Toolkit User Guide. Here, we provide instructions on how to use PaCS-Toolkit. If you encounter any issues and cannot be solved here, please feel free to submit a Github Issue. We hope PaCS-MD will be helpful for your research.

Content

What is PaCS-MD ?

PaCS-MD is an enhancement sampling method that exhaustedly samples the conformational space of given systems. PaCS-MD comprises multiple short MD simulations being cascaded by the selection of the starting structures for the next simulation cycles based on given evaluation criteria. By repeating this process, a large conformational space can be sampled without any mechanistic driven bias.

%%{ init: { 'flowchart': { 'curve': 'stepAfter'}, 'theme':'neutral', 'themeVariables': {'fontSize':'20px'} } }%%
flowchart TD;

cycle0["Preliminary MD simulations"];
ranking0["Structural Ranking"];
cycle0 --> ranking0;
ranking0 --> input_structures;

subgraph cycle["cycle"];
  ranking1["Structural Ranking"];
  judge{{"Fulfill requirements?"}};
  export1["Export top N snapshots to next cycle"];
  input_structures --> trajectories --> ranking1 --> judge;
  judge -- No --> export1 --> input_structures;
end

subgraph trajectories["Short MD simulations"];
  direction TB;
  traj1(trj1);
  traj2(trj2);
  etc_trj(...);
  traj(trj N);
end

subgraph input_structures[" "];
  direction TB;
  rank1(rank1);
  rank2(rank2);
  etc_sim(...);
  rank(rank N);
end

End((("finish")));
judge -- Yes --> End;

Citation

@article{Ikizawa2024,
  title = {PaCS-Toolkit: Optimized Software Utilities for Parallel Cascade Selection Molecular Dynamics (PaCS-MD) Simulations and Subsequent Analyses},
  volume = {128},
  ISSN = {1520-5207},
  url = {http://dx.doi.org/10.1021/acs.jpcb.4c01271},
  DOI = {10.1021/acs.jpcb.4c01271},
  number = {15},
  journal = {The Journal of Physical Chemistry B},
  publisher = {American Chemical Society (ACS)},
  author = {Ikizawa,  Shinji and Hori,  Tatsuki and Wijaya,  Tegar Nurwahyu and Kono,  Hiroshi and Bai,  Zhen and Kimizono,  Tatsuhiro and Lu,  Wenbo and Tran,  Duy Phuoc and Kitao,  Akio},
  year = {2024},
  month = apr,
  pages = {3631–3642}
}

Access

# Head
- Kitao

# Team leader
- Ikizawa

# Coding team including documentation
- Ikizawa (lead)
- Hori
- Kono

# Testing team
- Tegar (lead)
- Duy
- Bai
- Kimizono
- Lu