A New Machine Learning Approach Based on Range Corrected Deep Potential Model for Efficient Vibrational Frequency Computation

Vibrational spectrum simulation, as an ensemble average result, can be very time consuming when using high accuracy methods. Here, we introduce a new machine learning approach based on the range corrected deep potential (DPRc) model to improve computing efficiency. The approach was applied to computing \ch{C=O} stretching vibrational frequency shifts of formic acid-water solution. DPRc is adapted for frequency shift calculation. The system was divided into ``probe region'' and ``solvent region'' by atom. Three kinds of ``probe region'' were tested: single atom with atomic contribution correction, a single atom, and a single molecule. All data sets were prepared using by Quantum Vibration Perturbation (QVP) approach. The deep potential (DP) model was also adapted for frequency shift calculation for comparison, and different interaction cut-off radii were tested. The single molecule ``probe region'' results show the best accuracy, running roughly ten times faster than regular DP, while reducing the training time by a factor of about four, making it fully applicable in practice. The results show that dropping information of interaction distances between solvent atoms can significantly increase computing and training efficiency while ensuring little loss of accuracy. The protocol is practical, easy to apply, and extendable to calculating other physical quantities.

[1]  Carlos R. Baiz,et al.  Origin of thiocyanate spectral shifts in water and organic solvents. , 2022, The Journal of chemical physics.

[2]  M. Cho,et al.  Machine Learning Approach for Describing Water OH Stretch Vibrations. , 2021, Journal of chemical theory and computation.

[3]  Yaolong Zhang,et al.  Physically Motivated Recursively Embedded Atom Neural Networks: Incorporating Local Completeness and Nonlocality. , 2021, Physical review letters.

[4]  D. York,et al.  Development of Range-Corrected Deep Learning Potentials for Fast, Accurate Quantum Mechanical/Molecular Mechanical Simulations of Chemical Reactions in Solution. , 2021, Journal of chemical theory and computation.

[5]  Kristof T. Schütt,et al.  Machine learning of solvent effects on molecular spectra and reactions , 2020, Chemical science.

[6]  Jiali Gao,et al.  Quantum Vibration Perturbation Approach with Poly-atomic Probe in Simulating Infrared Spectra , 2021, Physical Chemistry Chemical Physics.

[7]  Wei Hu,et al.  A Machine Learning Protocol for Predicting Protein Infrared Spectra. , 2020, Journal of the American Chemical Society.

[8]  Yaolong Zhang,et al.  Efficient and Accurate Simulations of Vibrational and Electronic Spectra with Symmetry-Preserving Neural Network Models for Tensorial Properties. , 2020, The journal of physical chemistry. B.

[9]  Qiang Cui,et al.  Multi-level free energy simulation with a staged transformation approach. , 2020, The Journal of chemical physics.

[10]  Z. Guozhen,et al.  AI-based Spectroscopic Monitoring of Real-time Interactions between SARS-CoV-2 and Human ACE2 , 2020 .

[11]  M. Cho,et al.  Machine learning approach for describing vibrational solvatochromism. , 2020, The Journal of chemical physics.

[12]  Alexei A Kananenka,et al.  Machine Learning for Vibrational Spectroscopic Maps. , 2019, Journal of chemical theory and computation.

[13]  T. Elsaesser,et al.  Contact Ion Pairs of Phosphate Groups in Water - Two-Dimensional Infrared Spectroscopy of Dimethyl-Phosphate and Ab-Initio Simulations. , 2019, The journal of physical chemistry letters.

[14]  Jiali Gao,et al.  Active-Site Heterogeneity of Lactate Dehydrogenase , 2019, ACS Catalysis.

[15]  C. Bannwarth,et al.  GFN2-xTB-An Accurate and Broadly Parametrized Self-Consistent Tight-Binding Quantum Chemical Method with Multipole Electrostatics and Density-Dependent Dispersion Contributions. , 2018, Journal of chemical theory and computation.

[16]  E Weinan,et al.  End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems , 2018, NeurIPS.

[17]  Ana V. Cunha,et al.  Hydrophobic Collapse in N-Methylacetamide–Water Mixtures , 2018, The journal of physical chemistry. A.

[18]  Jiali Gao,et al.  Enhanced vibrational solvatochromism and spectral diffusion by electron rich substituents on small molecule silanes. , 2017, The Journal of chemical physics.

[19]  Jiali Gao,et al.  Perturbation Approach for Computing Infrared Spectra of the Local Mode of Probe Molecules. , 2017, Journal of chemical theory and computation.

[20]  M. Zanni,et al.  Watching Proteins Wiggle: Mapping Structures with Two-Dimensional Infrared Spectroscopy. , 2017, Chemical reviews.

[21]  J. Mondal,et al.  Water Dynamics in Gyroid Phases of Self-Assembled Gemini Surfactants. , 2016, Journal of the American Chemical Society.

[22]  Brian J. Smith,et al.  Structural Dynamics of a Catalytic Monolayer Probed by Ultrafast 2D IR Vibrational Echoes , 2011, Science.

[23]  Shaul Mukamel,et al.  Vibrational sum-frequency generation spectroscopy at the water/lipid interface: molecular dynamics simulation study. , 2010, Journal of the American Chemical Society.

[24]  Darius Abramavicius,et al.  Coherent Multidimensional Optical Spectroscopy of Excitons in Molecular Aggregates; Quasiparticle versus Supermolecule Perspectives , 2009 .

[25]  P. Mukherjee,et al.  Empirical amide I vibrational frequency map: application to 2D-IR line shapes for isotope-edited membrane peptide bundles. , 2009, The journal of physical chemistry. B.

[26]  M. Cho.  Coherent two-dimensional optical spectroscopy , 2006 .

[27]  R. Schweitzer‐Stenner.  Advances in vibrational spectroscopy as a sensitive probe of peptide and protein structure A critical review , 2006 .

[28]  J. Hirst,et al.  Modeling the amide I bands of small peptides. , 2006, The Journal of chemical physics.

[29]  S. Mukamel,et al.  Electrostatic DFT map for the complete vibrational amide band of NMA. , 2005, The journal of physical chemistry. A.

[30]  S. Ham,et al.  Correlation between electronic and molecular structure distortions and vibrational properties. II. Amide I modes of NMA–nD2O complexes , 2003 .

[31]  Jiali Gao,et al.  Hybrid Quantum and Molecular Mechanical Simulations: An Alternative Avenue to Solvent Effects in Organic Chemistry , 1996 .

[32]  Q. Du.  Vibrational spectroscopy of water interfaces , 1994 .

[33]  David C. Clary,et al.  Potential optimized discrete variable representation , 1992 .

[34]  D. O. Harris,et al.  Calculation of Matrix Elements for One‐Dimensional Quantum‐Mechanical Problems and the Application to Anharmonic Oscillators , 1965 .