Computational Phytochemistry

Satyajit Dey Sarker, Lutfun Nahar

• Publisher: Elsevier
• Year: 2018
• eBook ISBN: 9780128125465
• Topic: Physical Sciences
• Subtopic: Organic Chemistry

Computational Phytochemistry explores how recent advances in computational techniques and methods have been embraced by phytochemical researchers to enhance many of their operations, thus refocusing and expanding the possibilities of phytochemical studies. By applying computational aids and mathematical models to extraction, isolation, structure determination and bioactivity testing, researchers can extract highly detailed information about phytochemicals and optimize working approaches. This book aims to support and encourage researchers currently working with, or looking to incorporate, computational methods into their phytochemical work.Topics in this book include computational methods for predicting medicinal properties, optimizing extraction, isolating plant secondary metabolites and building dereplicated phytochemical libraries. The role of high-throughput screening, spectral data for structural prediction, plant metabolomics and biosynthesis are all reviewed, before the application of computational aids for assessing bioactivities and virtual screening are discussed. Illustrated with detailed figures and supported by practical examples, this book is an indispensable guide for all those involved with the identification, extraction and application of active agents from natural products.- Includes step-by-step protocols for various computational and mathematical approaches applied to phytochemical research- Features clearly illustrated chapters contributed by highly reputed researchers- Covers all key areas in phytochemical research, including virtual screening and metabolomics

Chapter 1

An Introduction to Computational Phytochemistry

Satyajit D. Sarker; Lutfun Nahar Liverpool John Moores University, Liverpool, United Kingdom

Abstract

This chapter provides an overview on ‘Computational Phytochemistry’, an emerging area of research, where computational tools, techniques and methods, artificial intelligence, and mathematical modelling are incorporated to resolve various issues in phytochemical research. In recent years, incorporation of computational techniques has been observed in screening plant materials, plant metabolomics, chemical fingerprinting, chemical taxonomy, biosynthesis and phylogenetic studies, prediction of pharmacological and toxicological properties (virtual screening or in silico studies), and automated structure determination of phytochemicals based on spectroscopic data. In most cases, the introduction of computer-aided approaches saves time and money associated with phytochemical research, ranging from bioactive compound discovery to identifying the metabolomes. In the coming years, ‘Computational Phytochemistry’ is set to transform the way we carry out phytochemical research today.

Keywords

Phytochemistry; Computation; Artificial intelligence; Virtual screening; Computer-aided methods; Metabolomics

Chapter Outline

• 1.1 Introduction
• 1.2 Computational Phytochemistry
• 1.3 Techniques, Theories, and Applications of Computational Phytochemistry
  • 1.3.1 Kohonen-Based Self-Organizing Map
  • 1.3.2 Density Functional Theory
  • 1.3.3 Docking Experiments and Virtual Screening (In Silico Screening)
  • 1.3.4 Structure Prediction and Structure Determination
  • 1.3.5 Chemometrics and Principal Component Analysis
  • 1.3.6 Data Mining and Databases
  • 1.3.7 Response Surface Methodology in Optimization of Extraction of Phytochemicals
  • 1.3.8 Computation in Isolation of Phytochemicals
  • 1.3.9 Miscellaneous
• 1.4 Conclusions
• References

1.1 Introduction

Computation is simply the act or process of computing, and the term ‘computational’ refers to any act or process relating to computation. Various degrees of computation are present all around us, and in almost everything we do. In fact, computation is no longer about machines, but about contributions of these machines to our lives in a globalized and interconnected world.

The tremendous advancement in computer science and its wide-ranging applications have influenced the way we carry out research (Sarker and Nahar, 2017). Computer has become an indispensable tool in research and development, as it is closely associated with analytical instrumentation and methods. It also serves as a tool for acquiring data, for word-processing and for handling electronic databases, and overall laboratory management and communication.

Chemistry has already incorporated computational techniques and mathematical modelling to address research questions and to develop new methods, which have eventually gave birth to a recognized branch in chemistry, known as ‘Computational Chemistry’. Simply, Computation Chemistry can be defined as a branch in chemistry that utilizes computer simulations to assist in solving chemical problems. It applies many methods of theoretical chemistry, embedded into efficient computer programmes, mainly to calculate the structures and properties of molecules and solids, e.g., electronic structure determination, geometry optimizations, frequency calculations, transition structures, docking, electron and charge distribution, rate constants, and many more (Sarker and Nahar, 2017). Gaussian 94, GAMESS, MOPAC, Spartan, and Sybyl are just a few of the popular software used in Computational Chemistry. Phenomenal technological progresses have led to a massive growth in the amounts of chemical data that are typically multivariate and tangled in structure (Bushkov et al., 2016). Therefore, several computational approaches have predominantly addressed dimensionality reduction and easy representation of multi-dimensional datasets to establish the relationships between the observed activity and calculated parameters commonly known as molecular descriptors. A molecular descriptor is the ultimate outcome of a logic and mathematical procedure that transforms chemical information encoded within a symbolic representation of a molecule into a useful number or the result of some standardized experiment (Todeschini and Consonni, 2000).

This chapter presents an overview on Computational Phytochemistry, an emerging area of research, where computational tools, techniques and methods, artificial intelligence, and mathematical modelling are incorporated to address various issues in phytochemistry and phytochemical research.

1.2 Computational Phytochemistry

Over the last few decades, noticeable increases in incorporation of computational techniques, artificial intelligence, and mathematical modelling in phytochemical research, especially in screening plant materials, plant metabolomics, chemical fingerprinting, chemical taxonomy, biosynthetic and phylogenetic studies, prediction of pharmacological and toxicological properties (virtual screening or in silico studies), and automated structure determination of phytochemicals based on spectroscopic data, have been observed (Sarker and Nahar, 2017). Some of these aspects initially formed ‘Phytochemical Informatics’ that dealt with large amounts of data related to phytochemicals and/or their sources (Ehrman et al., 2010), and this was probably the starting point of a new avenue in phytochemical research, now known as ‘Computational Phytochemistry’.

Computational Phytochemistry may be defined as an emerging branch of phytochemistry, where computational techniques and mathematical and statistical models are used to efficiently deal with various aspects of phytochemical research. Computational Phytochemistry, a product of the digital age, uses mathematical algorithms, statistics, and large databases to integrate theories and modelling with experimental observations. Creation of models and simulations of physical processes involved in phytochemical protocols, and application of statistics and data analysis techniques to extract useful information from large bodies of data, are two fundamental building blocks of Computational Phytochemistry.

In most cases, introduction of computer-aided approaches saves time and money associated with phytochemical research, ranging from bioactive compound discovery to identifying the metabolomes (Sarker and Nahar, 2017). The overall impact of computational methods on phytochemical research is already visible in recent publications, and this will steadily transform, over the coming years, the way we perform phytochemical research today.

There are several articles published on the use of computational approaches to solve a number of issues in phytochemical research (Nuzillard and Massiot, 1991; Stortz and Cerezo, 1992; Sumner et al., 2003; Rollinger et al., 2005; Cape et al., 2006; Desai and Gore, 2011; Jeeshna and Paulsamy, 2011; Barlow et al., 2012; Castellano et al., 2014; Ningthoujam et al., 2014; Das et al., 2017; Mocan et al., 2017), and relevant theories, useful methodologies and techniques have been presented there.

1.3 Techniques, Theories, and Applications of Computational Phytochemistry

1.3.1 Kohonen-Based Self-Organizing Map

Bushkov et al. (2016) utilized an artificial neural network incorporating the Kohonen-based self-organizing map (SOM) to study plant growth regulators, and it was the first example of a large-scale modelling in the field of agro-chemistry. Kohonen-based SOM was first introduced by the Finnish professor Teuvo Kohonen in the 1980s, and since then, it has been applied in many fields, especially in those which handle high-dimensional data sets.

The main objective of a SOM is to transform an incoming signal pattern of arbitrary dimension into a one- or two-dimensional discrete map and to perform this transformation adaptively in a topologically ordered fashion. Any SOM process has four major components: initialization, competition, cooperation, and adaptation. While in initialization all connection weights are initialized with small random values, in competition, for each input pattern, the neurons compute their respective values of a discriminant function that offers the basis for competition, where the neuron with the smallest value of the discriminant function wins. The winner neuron determines the spatial location of a topological neighbourhood of excited neurons and provides the basis for cooperation. In adaptation, the excited neurons decrease their individual values of the discriminant function in relation to the input pattern through suitable adjustment of the associated connection weig...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Dedication
• Contributors
• Foreword
• Preface
• Chapter 1: An Introduction to Computational Phytochemistry
• Chapter 2: Prediction of Medicinal Properties Using Mathematical Models and Computation, and Selection of Plant Materials
• Chapter 3: Optimization of Extraction Using Mathematical Models and Computation
• Chapter 4: Application of Computational Methods in Isolation of Plant Secondary Metabolites
• Chapter 5: Application of Computation in Building Dereplicated Phytochemical Libraries
• Chapter 6: High-Throughput Screening of Phytochemicals: Application of Computational Methods
• Chapter 7: Prediction of Structure Based on Spectral Data Using Computational Techniques
• Chapter 8: Application of Mathematical Models and Computation in Plant Metabolomics
• Chapter 9: Application of Computation in the Biosynthesis of Phytochemicals
• Chapter 10: Computational Aids for Assessing Bioactivities
• Chapter 11: Virtual Screening of Phytochemicals
• Index