This is a Python implementation of the molecular dynamics fingerprints (MDFP) methodology for building predictive models for phys-chem properties as delineated in our publication.
This toolkit is described and applied to the SAMPL6 logP prediction challenge, with the results published here.
Visit our documentation to learn details about installation, example workflow and API references.
Bibtex citations for the toolkit and the method are as follows:
@article{Wang2019,
doi = {10.1007/s10822-019-00252-6},
url = {https://doi.org/10.1007/s10822-019-00252-6},
year = {2019},
month = nov,
publisher = {Springer Science and Business Media {LLC}},
author = {Shuzhe Wang and Sereina Riniker},
title = {Use of molecular dynamics fingerprints ({MDFPs}) in {SAMPL}6 octanol{\textendash}water log P blind challenge},
journal = {Journal of Computer-Aided Molecular Design}
}
@article{Riniker2017,
doi = {10.1021/acs.jcim.6b00778},
url = {https://doi.org/10.1021/acs.jcim.6b00778},
year = {2017},
month = apr,
publisher = {American Chemical Society ({ACS})},
volume = {57},
number = {4},
pages = {726--741},
author = {Sereina Riniker},
title = {Molecular Dynamics Fingerprints ({MDFP}): Machine Learning from {MD} Data To Predict Free-Energy Differences},
journal = {Journal of Chemical Information and Modeling}
}
Copyright (c) 2018, shuzhe Wang
Project based on the Computational Molecular Science Python Cookiecutter