Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning
Abstract
Accurately predicting small molecule partitioning and hydrophobicity is critical in the drug discovery process. There are many heterogeneous chemical environments within a cell and entire human body. For example, drugs must be able to cross the hydrophobic cellular membrane to reach their intracellular targets, and hydrophobicity is an important driving force for drug–protein binding. Atomistic molecular dynamics (MD) simulations are routinely used to calculate free energies of small molecules binding to proteins, crossing lipid membranes, and solvation but are computationally expensive. Machine learning (ML) and empirical methods are also used throughout drug discovery but rely on experimental data, limiting the domain of applicability. We present atomistic MD simulations calculating 15,000 small molecule free energies of transfer from water to cyclohexane. This large data set is used to train ML models that predict the free energies of transfer. We show that a spatial graph neural network model achieves the highest accuracy, followed closely by a 3D-convolutional neural network, and shallow learning based on the chemical fingerprint is significantly less accurate. A mean absolute error of ~4 kJ/mol compared to the MD calculations was achieved for our best ML model. We also show that including data from the MD simulation improves themore »
- Authors:
-
- Biochemical and Biophysical Systems Group, Biosciences and Biotechnology Division, Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, California, United States
- Global Security Computing Applications, Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, California, United States
- Center for Applied Scientific Computing, Lawrence Livermore National Laboratory, 7000 East Avenue, Livermore, California, United States
- Publication Date:
- Research Org.:
- Lawrence Livermore National Laboratory (LLNL), Livermore, CA (United States)
- Sponsoring Org.:
- USDOE National Nuclear Security Administration (NNSA); USDOE Laboratory Directed Research and Development (LDRD) Program; American Heart Association (AHA)
- OSTI Identifier:
- 1657273
- Alternate Identifier(s):
- OSTI ID: 1729734
- Report Number(s):
- LLNL-JRNL-774697
Journal ID: ISSN 1549-9596
- Grant/Contract Number:
- AC52-07NA27344
- Resource Type:
- Published Article
- Journal Name:
- Journal of Chemical Information and Modeling
- Additional Journal Information:
- Journal Name: Journal of Chemical Information and Modeling; Journal ID: ISSN 1549-9596
- Publisher:
- American Chemical Society
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CHEMISTRY; Free energy; Interfaces; Molecular modeling; Molecules; Small molecules
Citation Formats
Bennett, W. F. Drew, He, Stewart, Bilodeau, Camille L., Jones, Derek, Sun, Delin, Kim, Hyojin, Allen, Jonathan E., Lightstone, Felice C., and Ingólfsson, Helgi I. Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning. United States: N. p., 2020.
Web. doi:10.1021/acs.jcim.0c00318.
Bennett, W. F. Drew, He, Stewart, Bilodeau, Camille L., Jones, Derek, Sun, Delin, Kim, Hyojin, Allen, Jonathan E., Lightstone, Felice C., & Ingólfsson, Helgi I. Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning. United States. https://doi.org/10.1021/acs.jcim.0c00318
Bennett, W. F. Drew, He, Stewart, Bilodeau, Camille L., Jones, Derek, Sun, Delin, Kim, Hyojin, Allen, Jonathan E., Lightstone, Felice C., and Ingólfsson, Helgi I. Fri .
"Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning". United States. https://doi.org/10.1021/acs.jcim.0c00318.
@article{osti_1657273,
title = {Predicting Small Molecule Transfer Free Energies by Combining Molecular Dynamics Simulations and Deep Learning},
author = {Bennett, W. F. Drew and He, Stewart and Bilodeau, Camille L. and Jones, Derek and Sun, Delin and Kim, Hyojin and Allen, Jonathan E. and Lightstone, Felice C. and Ingólfsson, Helgi I.},
abstractNote = {Accurately predicting small molecule partitioning and hydrophobicity is critical in the drug discovery process. There are many heterogeneous chemical environments within a cell and entire human body. For example, drugs must be able to cross the hydrophobic cellular membrane to reach their intracellular targets, and hydrophobicity is an important driving force for drug–protein binding. Atomistic molecular dynamics (MD) simulations are routinely used to calculate free energies of small molecules binding to proteins, crossing lipid membranes, and solvation but are computationally expensive. Machine learning (ML) and empirical methods are also used throughout drug discovery but rely on experimental data, limiting the domain of applicability. We present atomistic MD simulations calculating 15,000 small molecule free energies of transfer from water to cyclohexane. This large data set is used to train ML models that predict the free energies of transfer. We show that a spatial graph neural network model achieves the highest accuracy, followed closely by a 3D-convolutional neural network, and shallow learning based on the chemical fingerprint is significantly less accurate. A mean absolute error of ~4 kJ/mol compared to the MD calculations was achieved for our best ML model. We also show that including data from the MD simulation improves the predictions, tests the transferability of each model to a diverse set of molecules, and show multitask learning improves the predictions. This work provides insight into the hydrophobicity of small molecules and ML cheminformatics modeling, and our data set will be useful for designing and testing future ML cheminformatics methods.},
doi = {10.1021/acs.jcim.0c00318},
journal = {Journal of Chemical Information and Modeling},
number = ,
volume = ,
place = {United States},
year = {Fri Aug 14 00:00:00 EDT 2020},
month = {Fri Aug 14 00:00:00 EDT 2020}
}
https://doi.org/10.1021/acs.jcim.0c00318
Works referenced in this record:
Identification and Prediction of Promiscuous Aggregating Inhibitors among Known Drugs
journal, October 2003
- Seidler, James; McGovern, Susan L.; Doman, Thompson N.
- Journal of Medicinal Chemistry, Vol. 46, Issue 21
Sampling errors in free energy simulations of small molecules in lipid bilayers
journal, October 2016
- Neale, Chris; Pomès, Régis
- Biochimica et Biophysica Acta (BBA) - Biomembranes, Vol. 1858, Issue 10
Perspective on the Martini model
journal, January 2013
- Marrink, Siewert J.; Tieleman, D. Peter
- Chemical Society Reviews, Vol. 42, Issue 16
ADME Evaluation in Drug Discovery. 2. Prediction of Partition Coefficient by Atom-Additive Approach Based on Atom-Weighted Solvent Accessible Surface Areas
journal, May 2003
- Hou, T. J.; Xu, X. J.
- Journal of Chemical Information and Computer Sciences, Vol. 43, Issue 3
Fluorinated Alcohols’ Effects on Lipid Bilayer Properties
journal, August 2018
- Zhang, Mike; Peyear, Thasin; Patmanidis, Ilias
- Biophysical Journal, Vol. 115, Issue 4
Blind prediction of cyclohexane–water distribution coefficients from the SAMPL5 challenge
journal, September 2016
- Bannan, Caitlin C.; Burley, Kalistyn H.; Chiu, Michael
- Journal of Computer-Aided Molecular Design, Vol. 30, Issue 11
Basic ingredients of free energy calculations: A review
journal, January 2009
- Christ, Clara D.; Mark, Alan E.; van Gunsteren, Wilfred F.
- Journal of Computational Chemistry
Alcohol's Effects on Lipid Bilayer Properties
journal, August 2011
- Ingólfsson, Helgi I.; Andersen, Olaf S.
- Biophysical Journal, Vol. 101, Issue 4
Extended-Connectivity Fingerprints
journal, April 2010
- Rogers, David; Hahn, Mathew
- Journal of Chemical Information and Modeling, Vol. 50, Issue 5
Distribution of Amino Acids in a Lipid Bilayer from Computer Simulations
journal, May 2008
- MacCallum, Justin L.; Bennett, W. F. Drew; Tieleman, D. Peter
- Biophysical Journal, Vol. 94, Issue 9
The Properties of Known Drugs. 1. Molecular Frameworks
journal, January 1996
- Bemis, Guy W.; Murcko, Mark A.
- Journal of Medicinal Chemistry, Vol. 39, Issue 15
Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings
journal, January 1997
- Lipinski, Christopher A.; Lombardo, Franco; Dominy, Beryl W.
- Advanced Drug Delivery Reviews, Vol. 23, Issue 1-3
Lessons learned about steered molecular dynamics simulations and free energy calculations
journal, February 2019
- Boubeta, Fernando Martín; Contestín García, Rocío María; Lorenzo, Ezequiel Norberto
- Chemical Biology & Drug Design, Vol. 93, Issue 6
Drug metabolism and pharmacokinetics, the blood-brain barrier, and central nervous system drug discovery
journal, October 2005
- Alavijeh, Mohammad S.; Chishty, Mansoor; Qaiser, M. Zeeshan
- NeuroRX, Vol. 2, Issue 4
Advancing Drug Discovery through Enhanced Free Energy Calculations
journal, July 2017
- Abel, Robert; Wang, Lingle; Harder, Edward D.
- Accounts of Chemical Research, Vol. 50, Issue 7
Dissecting Machine-Learning Prediction of Molecular Activity: Is an Applicability Domain Needed for Quantitative Structure–Activity Relationship Models Based on Deep Neural Networks?
journal, November 2018
- Liu, Ruifeng; Wang, Hao; Glover, Kyle P.
- Journal of Chemical Information and Modeling, Vol. 59, Issue 1
PotentialNet for Molecular Property Prediction
journal, November 2018
- Feinberg, Evan N.; Sur, Debnil; Wu, Zhenqin
- ACS Central Science, Vol. 4, Issue 11
Small Molecule Hydration Free Energies in Explicit Solvent: An Extensive Test of Fixed-Charge Atomistic Simulations
journal, January 2009
- Mobley, David L.; Bayly, Christopher I.; Cooper, Matthew D.
- Journal of Chemical Theory and Computation, Vol. 5, Issue 2
Predicting a Drug’s Membrane Permeability: A Computational Model Validated With in Vitro Permeability Assay Data
journal, May 2017
- Bennion, Brian J.; Be, Nicholas A.; McNerney, M. Windy
- The Journal of Physical Chemistry B, Vol. 121, Issue 20
The MARTINI Coarse-Grained Force Field: Extension to Proteins
journal, April 2008
- Monticelli, Luca; Kandasamy, Senthil K.; Periole, Xavier
- Journal of Chemical Theory and Computation, Vol. 4, Issue 5
The MARTINI Force Field: Coarse Grained Model for Biomolecular Simulations
journal, July 2007
- Marrink, Siewert J.; Risselada, H. Jelger; Yefimov, Serge
- The Journal of Physical Chemistry B, Vol. 111, Issue 27
Molecular Dynamics Fingerprints (MDFP): Machine Learning from MD Data To Predict Free-Energy Differences
journal, April 2017
- Riniker, Sereina
- Journal of Chemical Information and Modeling, Vol. 57, Issue 4
In silico screening of drug-membrane thermodynamics reveals linear relations between bulk partitioning and the potential of mean force
journal, September 2017
- Menichetti, Roberto; Kanekal, Kiran H.; Kremer, Kurt
- The Journal of Chemical Physics, Vol. 147, Issue 12
Transfer of Arginine into Lipid Bilayers Is Nonadditive
journal, July 2011
- MacCallum, Justin L.; Bennett, W. F. Drew; Tieleman, D. Peter
- Biophysical Journal, Vol. 101, Issue 1
Calculating the free energy of transfer of small solutes into a model lipid membrane: Comparison between metadynamics and umbrella sampling
journal, October 2015
- Bochicchio, Davide; Panizon, Emanuele; Ferrando, Riccardo
- The Journal of Chemical Physics, Vol. 143, Issue 14
Statistical Convergence of Equilibrium Properties in Simulations of Molecular Solutes Embedded in Lipid Bilayers
journal, October 2011
- Neale, Chris; Bennett, W. F. Drew; Tieleman, D. Peter
- Journal of Chemical Theory and Computation, Vol. 7, Issue 12
Hydrophobicity scales: a thermodynamic looking glass into lipid–protein interactions
journal, December 2011
- MacCallum, Justin L.; Tieleman, D. Peter
- Trends in Biochemical Sciences, Vol. 36, Issue 12
Lead- and drug-like compounds: the rule-of-five revolution
journal, December 2004
- Lipinski, Christopher A.
- Drug Discovery Today: Technologies, Vol. 1, Issue 4
The COSMO and COSMO-RS solvation models: The COSMO and COSMO-RS solvation models
journal, April 2011
- Klamt, Andreas
- Wiley Interdisciplinary Reviews: Computational Molecular Science, Vol. 1, Issue 5
State of the art and prospects of methods for determination of lipophilicity of chemical compounds
journal, April 2019
- Kempińska, Dagmara; Chmiel, Tomasz; Kot-Wasik, Agata
- TrAC Trends in Analytical Chemistry, Vol. 113
Automatic Chemical Design Using a Data-Driven Continuous Representation of Molecules
journal, January 2018
- Gómez-Bombarelli, Rafael; Wei, Jennifer N.; Duvenaud, David
- ACS Central Science, Vol. 4, Issue 2
Molecular dynamics simulations and drug discovery
journal, October 2011
- Durrant, Jacob D.; McCammon, J. Andrew
- BMC Biology, Vol. 9, Issue 1
Characterizing the Chemical Space of ERK2 Kinase Inhibitors Using Descriptors Computed from Molecular Dynamics Trajectories
journal, May 2017
- Ash, Jeremy; Fourches, Denis
- Journal of Chemical Information and Modeling, Vol. 57, Issue 6
The Importance of Membrane Defects—Lessons from Simulations
journal, June 2014
- Bennett, W. F. Drew; Tieleman, D. Peter
- Accounts of Chemical Research, Vol. 47, Issue 8
Hydrophobic hydration from small to large lengthscales: Understanding and manipulating the crossover
journal, June 2005
- Rajamani, S.; Truskett, T. M.; Garde, S.
- Proceedings of the National Academy of Sciences, Vol. 102, Issue 27
A Method to Predict Blood-Brain Barrier Permeability of Drug-Like Compounds Using Molecular Dynamics Simulations
journal, August 2014
- Carpenter, Timothy S.; Kirshner, Daniel A.; Lau, Edmond Y.
- Biophysical Journal, Vol. 107, Issue 3