Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Machine Learning Models for Predicting Molecular UV–Vis Spectra with Quantum Mechanical Properties

Journal Article · · Journal of Chemical Information and Modeling

Accurate understanding of Ultraviolet–visible (UV–Vis) spectra is critical for highthroughput design of compounds for drug discovery. Experimentally determining UV–Vis spectra can become expensive when dealing with a large quantity of novel molecules. This provides us an opportunity to drive computational advances in molecular property predictions using quantum mechanics and machine learning. In this work, we use both Quantum Mechanically (QM) predicted and measured UV–Vis spectra as input to modify four different machine learning architectures: UVvis-SchNet, UVvis- DTNN, UVvis-Transformer, and UVvis-MPNN. Here we find that the UVvis-MPNN model outperforms the other models when using optimized 3D coordinates and QM predicted spectra as input features. This model has the highest performance for predicting UVVisible spectra with a training RMSE of 0.06 and validation RMSE of 0.08.

Research Organization:
Pacific Northwest National Laboratory (PNNL), Richland, WA (United States)
Sponsoring Organization:
USDOE; Defense Advanced Research Projects Agency (DARPA)
Grant/Contract Number:
AC05-76RL01830
OSTI ID:
1964168
Report Number(s):
PNNL-SA-171633
Journal Information:
Journal of Chemical Information and Modeling, Journal Name: Journal of Chemical Information and Modeling Journal Issue: 5 Vol. 63; ISSN 1549-9596
Publisher:
American Chemical SocietyCopyright Statement
Country of Publication:
United States
Language:
English

References (36)

Deep Learning Spectroscopy: Neural Networks for Molecular Excitation Spectra journal January 2019
In-Situ Microfluidic Study of Biphasic Nanocrystal Ligand-Exchange Reactions Using an Oscillatory Flow Reactor journal November 2017
Improved Infrared Spectra Prediction by DFT from a New Experimental Database journal April 2017
Progress and Challenges in the Calculation of Electronic Excited States journal September 2011
UV/Vis+ photochemistry database: Structure, content and applications journal September 2020
Anisotropy (optical, electrical and thermal conductivity) in thin polarizing films for UV/Vis regions of spectrum: Experimental and theoretical investigations journal March 2018
UV-adVISor: Attention-Based Recurrent Neural Networks to Predict UV–Vis Spectra journal November 2021
Computational Investigation of Drug Phototoxicity: Photosafety Assessment, Photo-Toxophore Identification, and Machine Learning journal October 2019
Machine Learning for Electronically Excited States of Molecules journal November 2020
Machine Learning Enables Highly Accurate Predictions of Photophysical Properties of Organic Fluorescent Materials: Emission Wavelengths and Quantum Yields journal February 2021
Predicting Infrared Spectra with Message Passing Neural Networks journal May 2021
SolTranNet–A Machine Learning Tool for Fast Aqueous Solubility Prediction journal May 2021
Big Data Meets Quantum Chemistry Approximations: The Δ-Machine Learning Approach journal April 2015
SchNetPack: A Deep Learning Toolbox For Atomistic Systems journal November 2018
Quantum Mechanical Methods Predict Accurate Thermodynamics of Biochemical Reactions journal March 2021
Extended-Connectivity Fingerprints journal April 2010
Deep Learning Optical Spectroscopy Based on Experimental Database: Potential Applications to Molecular Design journal March 2021
Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions
  • Marenich, Aleksandr V.; Cramer, Christopher J.; Truhlar, Donald G.
  • The Journal of Physical Chemistry B, Vol. 113, Issue 18, p. 6378-6396 https://doi.org/10.1021/jp810292n
journal May 2009
Quantum-chemical insights from deep tensor neural networks journal January 2017
Molecular excited states through a machine learning lens journal May 2021
Comparative dataset of experimental and computational attributes of UV/vis absorption spectra journal December 2019
Weaver's historic accessible collection of synthetic dyes: a cheminformatics analysis journal January 2017
Multi-fidelity prediction of molecular optical peaks with deep learning journal January 2022
COSMO: a new approach to dielectric screening in solvents with explicit expressions for the screening energy and its gradient journal January 1993
Self‐Consistent Molecular‐Orbital Methods. IX. An Extended Gaussian‐Type Basis for Molecular‐Orbital Studies of Organic Molecules journal January 1971
A consistent and accurate ab initio parametrization of density functional dispersion correction (DFT-D) for the 94 elements H-Pu journal April 2010
Self‐consistent molecular orbital methods. XXIII. A polarization‐type basis set for second‐row elements journal October 1982
Density‐functional thermochemistry. III. The role of exact exchange journal April 1993
NWChem: Past, present, and future journal May 2020
Deep learning for UV absorption spectra with SchNarc: First steps toward transferability in chemical compound space journal October 2020
Data Programs at NBS/NIST: 1901–2021 journal March 2022
PhotochemCAD 3: Diverse Modules for Photophysical Calculations with Multiple Spectral Databases journal February 2018
Using Extended-Connectivity Fingerprints with Laplacian-Modified Bayesian Analysis in High-Throughput Screening Follow-Up journal October 2005
Open Babel: An open chemical toolbox journal October 2011
The dye-sensitized solar cell database journal April 2018
Machine learning prediction of accurate atomization energies of organic molecules from low-fidelity quantum chemical calculations journal August 2019

Similar Records

Invariant Features for Accurate Predictions of Quantum Chemical UV-vis Spectra of Organic Molecules
Conference · Thu Feb 29 23:00:00 EST 2024 · OSTI ID:2351082

Predicting Energetics Materials’ Crystalline Density from Chemical Structure by Machine Learning
Journal Article · Mon Apr 26 00:00:00 EDT 2021 · Journal of Chemical Information and Modeling · OSTI ID:1784602

Learning together: Towards foundation models for machine learning interatomic potentials with meta-learning
Journal Article · Wed Jul 17 00:00:00 EDT 2024 · npj Computational Materials · OSTI ID:2403610