Metamorphosis of Computational Chemistry Driven by Artificial Intelligence and Industry 5.0

Garikapati Narahari Sastry is currently working as the Director of CSIR-North East Institute of Science and Technology, Jorhat. Prof. Sastry is a chemist, working in the interdisciplinary areas spanning chemistry, biology, modeling, and informatics. Prof. Sastry has been effective in employing computational and theoretical methods to solve problems in chemistry, biology, and allied areas. These efforts are embedded not only to provide a robust platform for carrying out research in CADD but also to inculcate the culture of developing software packages. He has made fundamental contributions in the areas of a) computational and theoretical chemistry; b) theoretical organic chemistry and reaction mechanism; c) software and data base development for drug discovery (Molecular Property Diagnostic Suite), d) non-covalent interactions, e) cooperativity of non-covalent interactions, f) computer-aided drug design. Under his guidance 29 people were awarded Ph.D., 20 Post-doctoral fellows, 235 students have done internship or short-term projects. In his career, he has delivered more than 480 lectures in international and national conferences/workshops/seminars. His research work was published in more than 330 research papers and reviews, which received over 12,554 citations, with an h-index of 55. Hridoy Jyoti Mahanta obtained a PhD in Computer Science and Engineering from Assam University, Silchar, India. He is currently working as a Scientist in the Advanced Computation and Data Sciences Division at CSIR-North East Institute of Science and Technology, Jorhat. His areas of interest include Artificial Intelligence, Machine Learning, Deep Learning, and their applications in the natural sciences, database development, and Software development. He has been closely working with Dr. G. Narahari Sastry for the past three and a half years, focusing on applying artificial intelligence and machine learning to solve fundamental problems in Bioinformatics, Chemoinformatics, and Chemistry. He has published around 36 papers in peer-reviewed journals and international conferences. Selvaraman Nagamani is a Scientist at Advanced Computation and Data Sciences Division, CSIR – North East Institute of Science and Technology, Jorhat, Assam, India. He obtained his PhD from Alagappa University, India and received the ICMR – Senior Research Fellowship (2014-2016). In 2017, he received prestigious DST – National Postdoctoral Fellowship to work with Dr. G. Narahari Sastry in CSIR – IICT Hyderabad. In 2021, he joined as a Scientist in Advanced Computation and Data Sciences Division, CSIR – NEIST. His research interests are developing open-source computational drug discovery software, applying novel and state-of-the-art computer aided drug design methods, network pharmacology, AI, and ML approaches in computational drug discovery. He has published more than 50 papers in peer review journals. Dinadayalane Tandabany has been Associate Professor of Chemistry at Clark Atlanta University, USA, since 2014. After being awarded his Ph.D in Chemistry from Pondicherry University, India, he took up a research position at Jackson State University, USA, where he conducted high performance computational investigations of structures, reactivities, electronic, transport and mechanical properties of carbon based nanomaterials, and taught a number of classes in general and computational chemistry prior to taking up his current role.He has co-authored over 70 papers and 8 book chapters, has been awarded a number of awards for his work, and has presented talks at numerous conferences. In addition, he actively works to help increase the number of underrepresented undergraduate and graduate students in computational chemistry and nanoscience research.

• 1. Artificial Intelligence1.1 A Comprehensive Introduction to AI1.2 Chemical Space and AI1.3 Impact of AI in Computational Chemistry1.4 Machine Learning Applications in Computational Chemistry1.4.1 ML and QSAR1.4.2 Property Prediction1.4.3 Generative Models for Molecular Design1.4.4 Chemical Reactions1.5 AI-Driven Approaches in Quantum Chemistry1.5.1 Quantum Property Prediction1.5.2 Quantum Circuit Optimization1.5.3 Quantum Machine Learning for Molecular Systems1.6 Future and Challenges of AI in ChemistryConclusionReferences2. Machine Learning2.1 Fundamental Concepts of Machine Learning2.2 Understanding the Foundations2.2.1 Human Learning and Types of Human Learning2.2.2 Human versus Machine Learning2.2.3 Supervised and Unsupervised Learning2.2.4 Neural Networks and Deep Learning2.2.5 Generative Learning2.3 Essential Steps in Applying Machine Learning2.3.1 Preparation and Handling of Data2.3.2 Feature Engineering and Feature Selection2.3.3 Model Building and Validation2.3.4 Evaluation2.3.5 Applicability Domain and Deployment2.4 Machine Learning in Various Fields of Natural Sciences2.4.1 Material Sciences2.4.2 Chemical Sciences2.4.3 Life Sciences2.4.4 Environmental Sciences2.4.5 Agricultural Sciences2.5 From Machine Learning to Deep Learning2.6 Rise of Generative Models and Industry 5.0ConclusionReferences3. Scientific Computing Using Python3.1 Python Basics3.1.1 Creating Python Environment3.1.2 Installation of Packages and Libraries3.1.3 Python Workbenches3.2 Handling Numeric Data with NumPy3.2.1 Basic Computing Operations3.2.2 Arrays and Indexing3.2.3 Vectorization3.2.4 Functions and Methods3.2.5 Dealing with Missing Data3.2.6 Generating Random Numbers3.3 Utilities of Pandas3.3.1 Reading and Handling Data with Pandas3.3.2 Selecting and Indexing3.3.3 Advanced Indexing3.3.4 Handling Text Data3.3.5 Statistical Functions with Pandas3.4 Visualization with Matplotlib and Seaborn3.4.1 Plotting Basics3.4.2 Various Types of Plots in Matplotlib3.4.3 Overlaying and Multifigure Plots3.4.4 3-Dimensional Plotting3.4.5 Seaborn Plot Types3.4.6 Categorical Plots3.4.7 Distribution and Pair Plots3.4.8 Correlation and Heatmaps3.5 RdKit for Chemoinformatics3.5.1 Molecule Representations3.5.2 Visualizing Structures3.5.3 Basic Operations with RdKit3.5.4 Finding Descriptors3.5.5 Atoms and Bonds3.5.6 Similarity and Searching Patterns3.6 Chemypy Package3.6.1 Basics and Installation3.6.2 Handling Reactions3.6.3 Chemical Kinetics3.6.4 Finding Properties3.6.5 Other UtilitiesConclusionReferences4. Machine Learning with Python4.1 Scikit-learn Library in Python4.2 Data Representation and Generation4.2.1 Generating Synthetic Data4.2.2 Exploring Data Sets4.2.3 Data Preprocessing and Preparation4.2.4 Feature Engineering4.3 Supervised Machine Learning4.3.1 Training for Linear Regression4.3.2 Multi-Linear Regression4.3.3 Classification with Logistic Regression4.3.4 Random Forest-Based Classification4.4 Unsupervised Machine Learning4.4.1 Dimensionality Reduction4.4.2 Clustering with Partitioned Algorithms4.4.3 Hierarchical Clustering4.5 Evaluation Metrics4.5.1 Confusion Matrix4.5.2 Accuracy and Error4.5.3 Precision and Recall4.5.4 ROC-AUC4.5.5 Cluster Analysis Metrics4.6 Case StudiesConclusionReferences5. Evolution of Computational Chemistry5.1 Overview of Computational Chemistry5.1.1 Computational Tools and Techniques5.1.2 Significance and Contributions5.2 Era of High-Performance Computing5.2.1 Role of Supercomputing in Computational Chemistry5.2.2 Parallelization and Acceleration Techniques5.2.3 Cloud Computing and Distributed Computing5.3 Software and Tools5.3.1 Overview of Computational Chemistry Software5.3.2 Open-Source vs. Commercial Software5.3.3 Popular Software Packages and Their Capabilities5.3.4 Intersection of Machine Learning and Computational Chemistry5.3.5 Predictive Modelling and Property Estimation5.4 Recent Advances and Future Directions5.4.1 Quantum Computing and Its Impact5.4.2 Multiscale Modelling and Simulation5.4.3 Emerging Fields and Interdisciplinary ApplicationsConclusionReferences6. Structure-Property Relationships6.1 Fundamentals of Structure-Property Relationships6.1.1 Defining Structure and Property6.1.2 Understanding Structure-Property Relationships6.1.3 Interlinking Molecular/Structural Features and Properties6.2 Chemical Structure and Property Correlations6.2.1 Molecular Structure and Properties6.2.2 Electronic Structure and Optical Properties6.2.3 Topological and Geometrical Descriptors6.3 Quantitative Structure-Property Relationships (QSPR)6.3.1 Developing QSPR Models6.3.2 Regression Analysis and Parameterization6.3.3 Applicability and Limitations6.4 Quantitative Structure-Activity Relationships (QSAR)6.4.1 QSAR in Drug Design6.4.2 Molecular Descriptors in QSAR6.4.3 Predictive Modelling and Toxicology6.5 Materials Science and Structure-Property Relationships6.5.1 Atomic and Crystal Structures6.5.2 Mechanical Properties of Materials6.5.3 Thermodynamic and Electronic Properties6.6 Biological Systems and Structure-Property Relationships6.6.1 Proteins and Enzymes6.6.2 DNA and RNA6.6.3 Structure-Function Relationships in BiologyConclusionReferences7. Reaction Modelling7.1 Overview of Reaction Modelling7.2 Chemical Kinetics7.2.1 Basics of Chemical Reactions7.2.2 Reaction Rate and Rate Laws7.2.3 Factors Affecting Reaction Rates7.3 Reaction Mechanisms7.3.1 Elementary Reactions vs. Overall Reactions7.3.2 Reaction Intermediates7.3.3 Reaction Mechanism Determination7.3.4 Analysis of Reaction Potential Energy Surface7.4 Reaction Rate Constants7.4.1 Arrhenius Equation7.4.2 Temperature Dependence7.4.3 Catalysis and Reaction Rate Constants7.5 Reaction Modelling Approaches7.5.1 Homogeneous vs. Heterogeneous Reactions7.5.2 Batch, Plug Flow, and Continuous Stirred Tank Reactors7.5.3 Ideal vs. Non-Ideal Reactors7.6 Numerical Methods for Reaction Modelling7.6.1 Finite Difference Methods7.6.2 Finite Element Methods7.6.3 Computational Fluid Dynamics (CFD)ConclusionReferences8. Computer-Aided Drug Design8.1 Introduction to Computer-Aided Materials (Drug) Design8.2 Drug Discovery Process8.2.1 Target Identification and Validation8.2.2 High-Throughput Screening (HTS)8.2.3 Hit-to-Lead Optimization8.2.4 Lead Optimization and Preclinical Testing8.3 Molecular Modeling in Drug Design8.3.1 Protein Structure Prediction8.3.2 Ligand Docking and Binding Affinity Prediction8.3.3 Pharmacophore Modeling8.3.4 Quantitative Structure-Activity Relationship (QSAR) Studies8.4 Virtual Screening and Compound Selection8.4.1 Structure-Based Virtual Screening8.4.2 Ligand-Based Virtual Screening8.4.3 Fragment-Based Drug Design8.5 De Novo Drug Discovery8.5.1 De Novo Molecular Design8.5.2 Computer-Generated Molecule Libraries8.5.3 Optimization Algorithms in Rational Drug Design8.6 Chemoinformatics and Bioinformatics8.6.1 Molecular Databases and Data Mining8.6.2 Sequence Analysis in Drug Discovery8.6.3 Chemoinformatics for Compound Analysis8.7 ADME/Toxicity Prediction8.7.1 Absorption, Distribution, Metabolism, and Excretion (ADME)8.7.2 Predicting Drug Toxicity8.7.3 Risk Assessment in Drug DesignConclusionReferences9. Materials Modelling9.1 Introduction9.1.1 Role of Materials Modelling in Science and Engineering9.1.2 Overview of Computational Methods9.2 Materials9.2.1 Predicting Mechanical Properties9.2.1.1 Strength and Elasticity9.2.1.2 Ductility and Toughness9.3 Material Design for Specific Applications9.3.1 Aerospace Materials9.3.2 Automotive Materials9.3.3 Building and Construction Materials9.4 Electronic and Photonic Materials9.4.1 Semiconductor Device Simulation9.4.1.1 Transistor Design9.4.1.2 Optoelectronic Device Modelling9.5 Superconductors and Magnetic Materials9.5.1 High-Temperature Superconductors9.5.2 Magnetic Data Storage Materials9.6 Energy Materials9.6.1 Fuel Cell and Battery Materials9.6.1.1 Lithium-Ion Batteries9.6.1.2 Fuel Cell Catalysts9.6.2 Solar Cell Materials9.6.2.1 Photovoltaic Device Optimization9.6.2.2 Organic Solar Cells9.7 Nanomaterials and Nanotechnology9.7.1 Modeling at the Nanoscale9.7.2 Nanoparticle Synthesis and Properties9.7.3 Nanocomposite MaterialsConclusionReferences10. Electronic Structure Calculation, Ab Initio, DFT, and MD Simulation10.1 Introduction to Quantum Mechanics10.1.1 Wave Functions, Probability Densities and Operators10.1.2 Postulates of Quantum Mechanics10.1.3 The Time-Independent Schrödinger Equation10.2 Molecular Hamiltonians and Operators10.2.1 Born-Oppenheimer Approximation10.2.2 Hamiltonian Operators10.2.3 Expectation Values and Observables10.3 Basis Sets and Wave Function Expansions10.3.1 Atomic Orbitals and Basis Functions10.3.2 Gaussian Basis Sets10.3.3 Slater-Type Orbitals (STOs)10.3.4 Types of Basis Sets10.3.5 Plane Waves and Fourier Transforms10.4 Introduction to Ab Initio Calculations10.4.1 Hartree-Fock Theory10.4.2 Configuration Interaction (CI)10.4.3 Many-Body Perturbation Theory (MBPT)10.5 Coupled Cluster Theory10.5.1 Cluster Operators and Excitations10.5.2 Single and Double Excitations (CCSD)10.5.3 Higher-Order Excitations (CCSD(T))10.6 Density Functional Theory (DFT)10.6.1 Hohenberg−Kohn Theorem10.6.2 Kohn-Sham Equations10.6.3 Local Density Approximation (LDA)10.6.4 Generalized Gradient Approximation (GGA)10.6.5 Dispersion Corrected DFT10.6.6 Hybrid Functional Theory10.7 Advanced Topics in DFT10.7.1 Time-Dependent DFT (TDDFT)10.7.2 Linear Response and Excitation Energies10.7.3 Optical Properties and Spectroscopy10.7.4 Beyond TDDFT: Nonlinear Response10.8 DFT for Strongly Correlated Systems10.8.1 Hubbard U and DFT+U10.8.2 Dynamical Mean Field Theory (DMFT)10.8.3 Quantum Monte Carlo and DFT+QMC10.9 Molecular Dynamics (MD) Simulation10.9.1 Introduction10.9.2 MD Using Simple Models10.9.3 MD with Continuous Potentials10.9.4 MD at Constant Temperatures and Pressures10.9.5 MD with Solvent Effects: Mean Force and Stochastic Dynamics10.9.6 Conformational Changes Post MD Simulation10.10 Quantum Mechanics/Molecular Mechanics (QM/MM)10.10.1 Hybrid Methods Overview10.10.2 Implementation and Applications10.10.3 Challenges and Considerations10.11 Advanced MD Techniques10.11.1 Umbrella Sampling10.11.2 Metadynamics10.11.3 Replica Exchange Molecular Dynamics (REMD)10.11.4 Ab Initio Molecular Dynamics (AIMD)10.12 Machine Learning in MD10.12.1 Force Field Parametrization10.12.2 Enhanced Sampling with ML10.12.3 Other Important Methods10.13 Simulation of Biomolecular Complexes10.13.1 Protein-Ligand Complexes10.13.2 Protein-Nucleic Acid Interactions10.13.3 Protein-Protein Interactions: Large Protein Assemblies and Their Interactions10.13.4 Membrane Proteins and Lipid Bilayers: Proteins Embedded in Lipid MembranesConclusionReferences11. The Chemical Space11.1 The Concept of Chemical Space11.2 Importance in Chemistry and Beyond11.3 The Chemical Spaces11.3.1 Chemical Space for Pharmacy11.3.2 Pharmacophore Space11.3.3 Polypharmacology11.3.4 Chemical Space for Natural Products11.3.5 Chemical Space for Biology and Medicine11.4 Docking for Virtual Screening of Chemical Space11.4.1 Different Open-Source Docking11.4.2 ML and DL Docking Tools for Chemical Space11.5 Chemoinformatic Resources for Chemical Space11.5.1 Various Chemical Space Databases11.5.2 Open-Source Platforms for Chemoinformatics11.5.3 Online Tools Developed for Mining Chemical and Target Spaces11.5.4 Useful Servers for Mining Chemical and Target Spaces of Target Families or Diseases11.6 Dimensions of Chemical Space11.6.1 Structural-Based Dimensions11.6.2 Descriptor-Based Dimensions11.7 Advanced Approaches to Explore the Chemical Space11.8 AI-ML Techniques and Tools for Chemical SpaceConclusionReferences12. Generative Models for Novel Catalyst Design12.1 Introduction12.2 Foundations of Catalyst Design12.2.1 Background and Significance of Catalyst Design12.2.2 The Evolution of Catalyst Development Approaches12.2.3 Role of Generative Models in Accelerating Catalyst Innovation12.2.4 Traditional Catalyst Discovery Methods and Limitations12.2.5 Need for Accelerated and Targeted Catalyst Design12.3 Generative Models in Chemistry12.3.1 Overview of Generative Models12.3.2 Machine Learning and Deep Learning in Chemistry12.3.3 Applications of Generative Models in Catalyst Design12.4 Types of Generative Models12.4.1 Variational Autoencoders (VAEs)12.4.2 Generative Adversarial Networks (GANs)12.4.3 Reinforcement Learning in Catalyst Design12.4.4 Comparative Analysis of Generative Model Types12.5 Catalyst Property Prediction12.5.1 Predictive Modeling for Catalyst Activity and Selectivity12.5.2 Quantitative Structure-Activity Relationship (QSAR) Models12.5.3 Challenges and Opportunities in Property Prediction12.6 Molecular Representation and Embedding12.6.1 Encoding Chemical Structures for Generative Models12.6.2 Graph Neural Networks (GNNs) in Catalyst Design12.6.3 Embedding Techniques for Catalyst Descriptor Generation12.7 Challenges and Considerations12.7.1 Data Quality and Bias in Training Datasets12.7.2 Transferability and Generalization of Generative Models12.7.3 Ethical Considerations in AI-Driven Catalyst Design12.8 Future Directions and Emerging Technologies12.8.1 Advanced Generative Models on the Horizon12.8.2 Integration with High-Throughput Experimentation12.8.3 Collaborative Approaches in Catalyst DesignConclusionReference13. Transforming Petroleum and Polymers Industry with AI13.1 Petrochemicals as Sustainable Materials for the Modern World13.2 Polymers13.2.1 Polymerization Process13.2.2 Polymer Synthesis from Petrochemicals13.2.3 Polymer Characterization13.2.4 Role of Catalyst in Polymerization13.3 Advanced Polymer Materials13.3.1 High-Performance Polymers for Specialized Applications13.3.2 Bio-Based and Sustainable Polymers13.3.3 Polymer Composites for Enhanced Properties13.4 Applying Machine Learning for Polymer Research13.4.1 Polymer Representations13.4.2 Generating New Polymer Chemistries13.4.3 Prediction of Properties for Sequence Defined Polymers13.4.4 Polymer Composites for Enhanced Properties13.4.5 Autonomous Experimentation13.5 Membrane Design for Petroleum Research13.5.1 Membrane Materials13.5.2 Membrane Types and Usage13.5.3 Revolutionizing Membrane Design Using Machine Learning13.5.4 Topology Optimization13.5.5 Predictive Models for Water Permeability and Salt Rejection13.6 Interpretable Discovery of Innovative Polymers and Membranes with AIConclusionReferences14. Application of Machine Learning and Artificial Intelligence in Natural Products Drug Discovery14.1 Introduction to Natural Products Drug Discovery14.1.1 Overview of Traditional Drug Discovery Methods14.1.2 Importance of Natural Products in Drug Development14.1.3 Challenges in Natural Products Drug Discovery14.2 Data Integration and Analysis14.2.1 Integration of Biological and Chemical Data14.2.2 Computational Approaches for Natural Products Screening14.2.3 Data Mining Techniques in Drug Discovery14.3 Predictive Modeling in Natural Products Research14.3.1 Predictive Modeling for Bioactivity14.3.2 Structure-Activity Relationship (SAR) Analysis14.3.3 QSAR (Quantitative Structure-Activity Relationship) Models14.4 Target Identification and Validation14.4.1 Identification of Drug Targets Using AI14.4.2 Validation of Targets in Natural Products Drug Discovery14.4.3 Challenges and Solutions in Target Identification14.5 Database Development for Natural Products14.5.1 Necessity for the Natural Products Databases14.5.2 NEIMPDB and OSADHI14.6 Prediction of Targets and Biological Activity of Natural Products14.7 Visualizing and Navigating the Natural Products Space in Chemical Space14.8 The Natural Product Database LandscapeConclusionReferences15. Applying Machine Learning in Drug Repurposing15.1 Drug Repurposing15.1.1 Overview of Traditional Drug Development15.1.2 Rationale for Drug Repurposing15.2 Role of Machine Learning and Artificial Intelligence in Drug Repurposing15.2.1 Evolution of AI in Drug Discovery15.2.2 Machine Learning Approaches in Repurposing15.2.3 Applications of AI in Identifying Repurposable Drugs15.2.4 Data Integration and Analysis for Drug Repurposing15.3 Integration of Biomedical Data Sources15.3.1 Computational Approaches for Data Analysis15.3.2 Utilizing Real-World Evidence (RWE) in Repurposing15.4 Network Pharmacology and Drug Repurposing15.4.1 Network-Based Approaches to Drug Repurposing15.4.2 Analysis of Biological Pathways and Networks15.4.3 Predictive Modeling in Network Pharmacology15.5 Predictive Analytics for Drug Repurposing15.5.1 Predictive Modeling Techniques15.5.2 Machine Learning Algorithms for Prediction15.5.3 Quantitative Structure-Activity Relationship (QSAR) in Repurposing15.6 High-Throughput Screening and Virtual Screening in Drug Repurposing15.6.1 Automation in High-Throughput Screening15.6.2 Virtual Screening Using AI Algorithms15.6.3 Integration of Experimental and Computational Screening15.7 Identification of Novel Targets for Drug Repurposing15.7.1 Target Identification Using AI15.7.2 Validation of Targets for Repurposing15.7.3 Challenges and Strategies in Target Identification15.8 Combination Therapy and Synergistic Drug Repurposing15.8.1 AI-Guided Combination Therapy Strategies15.8.2 Identifying Synergistic Drug Combinations15.8.3 Challenges and Opportunities in Combination Repurposing15.9 Ethical and Regulatory Considerations in Drug Repurposing15.9.1 Ethical Issues in AI-Driven Drug Repurposing15.9.2 Regulatory Compliance and Safety Assessment15.9.3 Balancing Innovation with Ethical and Legal Frameworks15.10 Implications for the Future of Drug Repurposing with AIConclusionReferences16. From Industry 4.0 to Industry 5.0: The Role of AI and Computational Chemistry16.1 Introduction16.2 Fourth Industrial Revolution and the Rise of Industry 5.016.3 Role of Artificial Intelligence (AI) in Industry 5.016.4 Towards an AI-Assisted, Automated Chemistry Lab16.4.1 Robotics and AI16.4.2 Concept of the "Self-Driving" Lab16.4.3 Remaining Hurdles to Realize the Vision of Industry 5.016.4.4 Will AI Ever Replace Human Chemical Intuition?16.5 AI in Industry 5.0: Driving Smart Manufacturing16.5.1 The Role of AI in Predictive Maintenance and Quality Control16.5.2 AI-Powered Robotics and Automation16.5.3 AI-Driven Supply Chain Optimization16.5.4 AI-Driven Materials Databases and Repositories16.6 Challenges and Opportunities of Industry 5.016.7 Future Trends in Reaction Modelling16.7.1 Advances in Computational Approaches16.7.2 Integration of Machine Learning and AI16.7.3 Emerging Applications and Challenges16.7.4 Importance of Reaction Modelling in Science and Industry16.8 Machine Learning for Materials Simulation16.8.1 High Throughput Virtual Screening16.8.2 Analyzing Experimental Data16.8.3 Active Learning for Materials16.9 Various Tools for Computational Chemistry Using AI16.10 Evolution of Chemical and Biological Space Using AI16.11 Open-Source Tools of Computational Chemistry Using AI, ML, and DL16.12 Latest Interventions of AI in Computational Chemistry16.13 Future ProspectsConclusionReferences