Skip to main content

Deep-Learning Driven Adaptive Molecular Simulations

Project description

DeepDriveMD-F (DeepDriveMD-pipeline)

DeepDriveMD-F: Deep-Learning Driven Adaptive Molecular Simulations (file-based continual learning loop)

Documentation Status

Details can be found in the ducumentation. For more information, please see our website.

How to run

Setup

Install deepdrivemd into a virtualenv with:

python3 -m venv env
source env/bin/activate
pip install --upgrade pip setuptools wheel
pip install -e .

Then, install pre-commit hooks: this will auto-format and auto-lint on commit to enforce consistent code style:

pre-commit install
pre-commit autoupdate

In some places, DeepDriveMD relies on external libraries to configure MD simulations and import specific ML models.

For MD, install the mdtools package found here: https://github.com/braceal/MD-tools

For ML (specifically the AAE model), install the molecules package found here: https://github.com/braceal/molecules/tree/main

Generating a YAML input spec:

First, run this command to get a sample YAML config file:

python -m deepdrivemd.config

This will write a file named deepdrivemd_template.yaml which should be adapted for the experiment at hand. You should configure the molecular_dynamics_stage, aggregation_stage, machine_learning_stage, model_selection_stage and agent_stage sections to use the appropriate run commands and environment setups.

Running an experiment

Then, launch an experiment with:

python -m deepdrivemd.deepdrivemd -c <experiment_config.yaml>

This experiment should be launched

Note on input data

The input PDB and topology files should have the following structure:

ls data/sys*

data/sys1:
comp.pdb comp.top

data/sys2:
comp.pdb comp.top

Where the topology files are optional and only used when molecular_dynamics_stage.task_config.solvent_type is "explicit". Only one system directory is needed but an arbitrary number are supported. Also note that the system directory names are arbitrary. The path to the data directory should be passed into the config via molecular_dynamics_stage.initial_pdb_dir.

MIT License

Copyright (c) 2021 DeepDriveMD-F

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

deepdrivemd-0.0.1.tar.gz (32.9 kB view hashes)

Uploaded Source

Built Distribution

deepdrivemd-0.0.1-py3-none-any.whl (43.5 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page