Abstract
Naturally occurring compounds are found to be the most prominent and
effective biological active compounds against various diseases. The majority of drugs
approved between 1983 to 1994 are derived from natural products. Still today, the
majority of pharmaceutical laboratories are hoping to get new drug candidates from
natural resources. The traditional method of drug discovery from naturally occurring
compounds has been upgraded by using advanced computer-based drug discovery.
In drug discovery, the initial efforts are to know the relationship between the biological
activity of natural compounds and their chemical structures. To be precise, the method
of structure-activity relationship aims to recognize the basic structural component
responsible for biological activity.
The computational modeling drug discovery using various tools plays a major role in
identifying the lead compounds. In this method, three major ways are utilized to
understand the structure-activity relationship.
The foremost one is the Quantitative Structure-Activity Relationship (QSAR). In this
method, the relationship was established using regression techniques between the
‘Predictor Variable (X)’ with the potency of the ‘Response Variable (Y)’. The predictor
variables are molecular descriptors, while the response variables represent the
biological activities of the molecules against the selected diseases. If the response
variable represents the chemical property, in that case, the model is called as
Quantitative Structure-Property Relationship (QSPR).
The second method is called “Inhibition Studies”. In this process, the designed
chemical entity is docked to the targeted enzyme using docking software. The basic
principle of this method is the executive competitive inhibition between the natural
inhibitor and the designed chemical entity. The law of thermodynamic is used to
understand the best-docked chemical entity by obtaining the value of binding energy
(ΔG kcal/mole) due to the complex formation between the chemical moiety and target
enzyme.
The third approach is very advanced and more accurate. It is called “The drug
This chapter discussed all three methods in detail, along with examples. It also provides
The final aim of this chapter is not only to provide the theoretical background of drug
The third approach is very advanced and more accurate. It is called “The drug
discovery using Artificial Neural Network”. This is the recent technique adapted by
major international pharmaceutical research laboratories. In this method, the neural
network is designed and trained to identify the potent chemical compound against a
particular disease. The designing of the network can be achieved using the chemical
properties of a neuron, and output is related to the biological activity.
This chapter discussed all three methods in detail, along with examples. It also provides
the practical procedure to use available computational tools.
The final aim of this chapter is not only to provide the theoretical background of drug
discovery using structure-activity relationships but also to provide practical methods.