Atom Modeling PipeLine

RESOURCE

Abstract

This repository contains code for the computational framework to build ML models for drug discovery purposes. Our end-to-end pipeline extracts features from model-ready datasets and trains and saves the model to our model zoo or to disk. Our pipeline generates a variety of molecular features and both shallow and deep ML models. The HPC-specific module we have developed conducts efficient parallelized search of the model hyperparameter space and reports the best-performing hyperparameters for each of these feature/model combinations.
Developers:
Minnich, Amanda [1] McLoughlin, Kevin [1] Allen, Jonathan [1] Forsyth, Ryan [1]
  1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Contributors:
Researcher: Weber, Claire [1] Murad, Neha [1] Tse, Margaret [1] Weber, Andrew [1] Deng, Jason [1] Madej, Ben [2]
  1. GlaxoSmithKline plc (GSK)
  2. Frederick National Laboratory
Contributing Organizations:
Research Group: Frederick National Laboratory for Cancer Research
Research Group: GlaxoSmithKline plc (GSK)
Release Date:
2019-10-26
Project Type:
Open Source, Publicly Available Repository
Software Type:
Scientific
Version:
1.0
Licenses:
MIT License
Sponsoring Org.:
Code ID:
32934
Site Accession Number:
997148
Research Org.:
Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
Country of Origin:
United States

RESOURCE

Citation Formats

Minnich, Amanda J., McLoughlin, Kevin S., Allen, Jonathan E., Forsyth, Ryan M., Weber, Claire, Murad, Neha, Tse, Margaret, Weber, Andrew, Deng, Jason, and Madej, Ben. Atom Modeling PipeLine. Computer Software. https://github.com/ATOMconsortium/AMPL. USDOE National Nuclear Security Administration (NNSA). 26 Oct. 2019. Web. doi:10.11578/dc.20191211.1.
Minnich, Amanda J., McLoughlin, Kevin S., Allen, Jonathan E., Forsyth, Ryan M., Weber, Claire, Murad, Neha, Tse, Margaret, Weber, Andrew, Deng, Jason, & Madej, Ben. (2019, October 26). Atom Modeling PipeLine. [Computer software]. https://github.com/ATOMconsortium/AMPL. https://doi.org/10.11578/dc.20191211.1.
Minnich, Amanda J., McLoughlin, Kevin S., Allen, Jonathan E., Forsyth, Ryan M., Weber, Claire, Murad, Neha, Tse, Margaret, Weber, Andrew, Deng, Jason, and Madej, Ben. "Atom Modeling PipeLine." Computer software. October 26, 2019. https://github.com/ATOMconsortium/AMPL. https://doi.org/10.11578/dc.20191211.1.
@misc{ doecode_32934,
title = {Atom Modeling PipeLine},
author = {Minnich, Amanda J. and McLoughlin, Kevin S. and Allen, Jonathan E. and Forsyth, Ryan M. and Weber, Claire and Murad, Neha and Tse, Margaret and Weber, Andrew and Deng, Jason and Madej, Ben},
abstractNote = {This repository contains code for the computational framework to build ML models for drug discovery purposes. Our end-to-end pipeline extracts features from model-ready datasets and trains and saves the model to our model zoo or to disk. Our pipeline generates a variety of molecular features and both shallow and deep ML models. The HPC-specific module we have developed conducts efficient parallelized search of the model hyperparameter space and reports the best-performing hyperparameters for each of these feature/model combinations.},
doi = {10.11578/dc.20191211.1},
url = {https://doi.org/10.11578/dc.20191211.1},
howpublished = {[Computer Software] \url{https://doi.org/10.11578/dc.20191211.1}},
year = {2019},
month = {oct}
}