New Horizons in Computational Chemistry Software (Topics in Current Chemistry Collections)

Description

This volume presents the current status of software development in the field of computational and theoretical chemistry and gives an overview of the emerging trends. The challenges of maintaining the legacy codes and their adaptation to the rapidly growing hardware capabilities and the new programming environments are surveyed in a series of topical reviews written by the core developers and maintainers of the popular quantum chemistry and molecular dynamics programs. Special emphasis is given to new computational methodologies and practical aspects of their implementation and application in the computational chemistry codes. Modularity of the computational chemistry software is an emerging concept that enables to bypass the development and maintenance bottleneck of the legacy software and to customize the software using the best available computational procedures implemented in the form of self-contained modules. Perspectives on modular design of the computer programs for modeling molecular electronic structure, non-adiabatic dynamics, kinetics, as well as for data visualization are presented by the researchers actively working in the field of software development and application. This volume is of interest to quantum and computational chemists as well as experimental chemists actively using and developing computational software for their research. Chapters "MLatom 2: An Integrative Platform for Atomistic Machine Learning” and “Evolution of the Automatic Rhodopsin Modeling (ARM) Protocol" are available open access under a CC BY 4.0 License via link.springer.com. Contents Preface Technological Advances in Remote Collaborations Abstract 1 Introduction 2 Education 3 Collaboration Tools 4 Community 5 Virtual Conferences 6 Future Outlook 7 Conclusion Acknowledgements References MLatom 2: An Integrative Platform for Atomistic Machine Learning Abstract 1 Introduction 2 Overview 2.1 ML Tasks 2.1.1 Using ML Models 2.1.2 Creating ML Models 2.1.3 Estimating Accuracy of ML Models 2.1.4 Multi-step Tasks 2.1.5 Learning Curves 2.1.6 ML Nuclear Ensemble Spectra 2.2 Data Set Tasks 2.2.1 Splitting and Sampling 2.2.2 Analysis of Data 3 Native Implementations 3.1 Kernel Ridge Regression 3.2 KREG 3.3 Coulomb Matrix 3.4 Permutationally Invariant Kernel 4 Interfaces 4.1 Hyperopt 4.2 sGDML 4.3 GAP-QUIP 4.4 TorchANI 4.5 DeePMD-kit 4.6 PhysNet 5 Applications 5.1 Case Study 1: Hyperparameter Optimization 5.2 Case Study 2: Learning Curves 5.3 Case Study 3: Δ-Learning and Structure-Based Sampling 5.4 Case Study 4: Absorption Spectrum 6 Conclusions References Reaction Space Projector (ReSPer) for Visualizing Dynamic Reaction Routes Based on Reduced-Dimension Space Abstract 1 Introduction 2 Reaction Space Projector 2.1 Pre-Processing for CMDS 2.2 Classical Multidimensional Scaling 2.3 Recent Applications of ReSPer 3 Applications to Au5 Cluster 3.1 Reduced-Dimensionality Reaction Space and Potential Energy Landscape 3.2 Reaction Dynamics of Isomerization Reaction on Two-Dimensional Reaction Space 3.3 Closed Islands: Linkage of NPI Isomers of MIN1 3.4 A Bundle of Trajectories Related to Bifurcation Reaction 4 Conclusion Acknowledgements References NAST: Nonadiabatic Statistical Theory Package for Predicting Kinetics of Spin-Dependent Processes Abstract 1 Introduction 2 Nonadiabatic Statistical Theory 2.1 Microcanonical Rate Constants 2.2 Microcanonical Transition Probabilities 2.3 Canonical Rate Constants 2.4 Velocity-Averaged Probabilities 2.5 Rate Constants and Transition Probabilities Between Individual MS Components of Spin States 2.6 Transition State Theory Rate Constants 2.7 Effective Hessian 3 NAST Package Capabilities and Implementation 3.1 Forward and Reverse Rate Constants 3.2 Transition Probabilities 3.3 Rate Constants and Transition Probabilities Between Individual MS Components of Spin States 3.4 Rate Constants in Solution 3.5 Transition State Theory Rate Constants 3.6 Effective Hessian Tool effhess 3.7 IRC Fitting Tool ircfit 3.8 Modular Structure of the NAST Package 4 Examples of Applications 4.1 Isomerization of Propylene Oxide to Acetone and Propanal 4.2 Spin-Forbidden Isomerization of Ni(dpp)Cl2 4.3 T1 S0 Relaxation in Cyclopropene 5 Conclusions Acknowledgements References Evolution of the Automatic Rhodopsin Modeling (ARM) Protocol Abstract 1 Introduction: Contents and Scope 2 Rhodopsins: a Family of Biological Photoreceptors 2.1 Structure and Diversity 2.2 Biological Functions 2.3 Photoreactivity 2.4 Applications of Natural and Engineered Rhodopsins: Optogenetics 3 The Original Version of ARM: a Pioneer Technology for Rhodopsin QMMM Modeling 3.1 State-of-the-Art for QMMM Modeling of Rhodopsins 3.2 ARM Scope 3.3 Definition of an ARM QMMM Model 3.4 QMMM Model Generator 3.4.1 Classical Molecular Dynamics Simulations 3.4.2 QMMM Calculations 3.5 Automation Issues 4 a-ARM: the First Major Update Towards Automation 4.1 Methodological Aspects 4.2 Software Implementation Aspects 4.3 Benchmark, Validation and Application Aspects 4.4 Limitations and Pitfalls of a-ARM 4.5 Recent Updates and Improvements 5 Web-ARM, a Web-Based Interface to ARM 5.1 Interface Features 5.2 Limitations and Future Development of Web-ARM 6 PyARM 6.1 Package Description 6.2 Current ARM-Based QMMM Protocols 6.3 PyARM Technical Details 6.3.1 PyARM Default Parameters 6.3.2 PyARM Installation 6.3.3 PyARM Tailoring 6.4 Color Tuning Analysis in Terms of Steric and Electrostatic Effects 6.4.1 Computed Quantities 6.4.2 Steric Effects 6.4.3 Electrostatic Effects 6.4.4 Turn-Off Module 6.5 PyARM Current Accuracy and Drawbacks 7 Outlook and Concluding Remarks Acknowledgements References Coupled- and Independent-Trajectory Approaches Based on the Exact Factorization Using the PyUNIxMD Package Abstract 1 Introduction 2 Exact Factorization 3 Mixed Quantum-Classical Approaches Based on the Exact Factorization 3.1 Coupled-Trajectory Approach 3.2 Independent-Trajectory Approach 3.3 Available Programs 4 The PyUNIxMD Package 4.1 Interfacing with Quantum Chemistry Programs 4.2 Procedures for ESMD Simulations with PyUNIxMD 5 Numerical Results 6 Conclusion Acknowledgements References The Static–Dynamic–Static Family of Methods for Strongly Correlated Electrons: Methodology and Benchmarking Abstract 1 Introduction 2 The SDS Family of Methods 2.1 SDSCI, SDSPT2, iCI, iCIPT2, and iVI 2.2 iCAS and iCISCF(2) 3 Benchmarking SDSCI and SDSPT2 3.1 Choice of  in SDSCI and SDSPT2 3.2 Results and Discussion 3.2.1 Size Consistency 3.2.2 Excited States of Closed-Shell Systems 3.2.3 Statistical Analysis of Singlet States 3.2.4 Statistical Analysis of Triplet States 3.3 Excited States of Open-Shell Systems 4 Conclusions and Outlook Acknowledgements References Ensemble Density Functional Theory of Neutral and Charged Excitations Abstract 1 Introduction 2 Unified Ensemble DFT Formalism for Neutral and Charged Excitations 2.1 DFT of Neutral Excitations 2.1.1 GOK Ensembles 2.1.2 DFT of GOK Ensembles 2.1.3 Extraction of Individual State Properties 2.2 DFT of Charged Excitations: N-Centered Ensemble Formalism 2.2.1 N-Centered Ensembles 2.2.2 DFT of N-Centered Ensembles 2.2.3 Exact Ionization Potential and Electron Affinity Theorems 3 Equivalence Between Weight Derivatives and xc Derivative Discontinuities 3.1 Review of the Regular PPLB Approach to Charged Excitations 3.1.1 Ensemble Formalism for Open Systems 3.1.2 DFT for Fractional Electron Numbers 3.1.3 Kohn–Sham PPLB 3.1.4 Janak’s Theorem and Its Implications 3.1.5 Fundamental Gap Problem 3.1.6 Exchange-Only Derivative Discontinuity 3.2 Connection Between PPLB and N-Centered Pictures 3.3 Suppression of the Derivative Discontinuity 4 The Exact Hartree-Exchange Dilemma in eDFT 4.1 Extending the HF Method to Ensembles 4.1.1 Ensemble Density Matrix Functional Approach 4.1.2 Ghost Interaction Errors 4.1.3 State-Averaged HF Approach 4.1.4 eDMHF Versus SAHF 4.2 Concavity of Approximate Energies and Lieb Maximization 4.3 Insights from the Hubbard Dimer Model 4.3.1 SAHF and eDMHF Energy Expressions 4.3.2 Symmetric Case 4.3.3 Single SAHF Ensemble v-Representability Issue 4.4 Exact Self-Consistent eDFT Based on SAHF 4.5 Connection with Practical Hybrid eDFT Calculations 5 Individual Correlations within Ensembles: An Exact Construction 5.1 State-of-the-Art Ensemble Correlation DFAs and Beyond 5.2 Weight Dependence of the KS Wave Functions in GOK-DFT 5.3 Extraction of Individual Correlation Energies 5.4 Individual Correlations Versus Individual Components 5.5 Density-Driven Ensemble Correlation Energy Expression 5.6 Application to the Hubbard Dimer 5.6.1 Exact Theory and Approximations 5.6.2 Results and Discussion 6 Conclusions and Perspectives Acknowledgements References