Drug Design Is An Inventive Process Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Drug design is an inventive process of finding new medications based on the knowledge of the biological target also known as rational drug design. It involves designing of smaller new molecules which complementary in shape and charge to the biomolecular target to which they interact and bind to it13. Drug design is frequently based on computer modeling techniques which is often referred to as computer-aided drug design.


Now-a-days in the field of new drug discovery and development , computational techniques are rapidly gaining popularity, implementation and appreciation. Different terms are being applied to this area, including computer-aided drug design (CADD), computational drug design, computer-aided molecular design (CAMD), computer-aided molecular modeling (CAMM), rational drug design, insilico drug design, computer-aided rational drug design. Both computational and experimental techniques have complementary roles in drug discovery and development. 13, 14

CADDD entails:

Drug discovery and development process is streamlined by using the computing power.

Chemical and biological information about ligands, targets in process of identification and optimization of new analogs.

Design of  insilico filters to eliminate compounds with undesirable properties (ADMET) and select the most promising candidates.

Two types of drug design are as follows 1. Ligand-based drug design 

2. Structure-based drug design.

Ligand based

Ligand-based drug design (or indirect drug design) depends on knowledge of all other molecules which binds to the biological target. Knowledge based molecules used to derive a pharmacophore model possessing necessary structural characteristics in order to bind to the target.  Alternatively, QSAR study was used to predict the activity of new analogs by correlating the calculated properties of molecules and their experimentally determined biological activity .13,14

Structure based

Structure-based drug design (or direct drug design) mainly depends on knowledge of the three dimensional structure of the biological target .3D structure of targets are obtained using the x-ray crystallography or NMR spectroscopy techniques which is used to predict the activity of candidate molecules based on their high affinity and selectivity to the target14 .Alternatively various automated computational procedures may be used to suggest new drug candidates.

Flow charts of two strategies of a)structure b)ligand based drug design




Pharmacophore means "a set of structural features in a molecule that is recognized at a receptor site and is responsible for that molecule's biological activity"

Pharmacophore modeling studies have become one of the major tools in field of drug discovery .Historically, pharmacophores were established by Lemont Kier, who first mentioned the concept in 1967 but used the term in publication in 1971. In 1909, Paul Ehrlich introduced the concept of pharmacophore, who defined the pharmacophore as 'a molecular framework that carries (phoros) the essential features responsible for a drug's (pharmacon) biological activity'15.

The IUPAC defines  "A pharmacophore is ensemble of steric and electronic features that is necessary to ensure the optimal supramolecular interactions with a specific biological target and to trigger (or block) its biological response."16 A pharmacophore is an abstract description of molecular features which are necessary for  molecular recognition of a ligand by a biological macromolecule.

Pharmacophore features

Hydrogen bond acceptor

Hydrogen bond donor


Hydrophobic aliphatic

Hydrophobic aromatic

Positive ionizable

Negative ionizable

Ring aromatic

In order to identify novel ligands, the pharmacophoric features should match different chemical moities with similar properties. A well-defined pharmacophore model includes both hydrophobic volumes and hydrogen bond vectors.

Various ligand-based and structure-based methods involving pharmacophore modeling have been developed and extensively applied in field of virtual screening, de novo design and lead optimization.

Framework of Pharmacophore architecture.


Pharmacophore identification process

Identification of a pharmacophore17 involves two steps:

1. Analyzing the molecules to identify pharmacophoric features, that is, atoms that can interact with a receptor.

2. Aligning the active conformations of the molecules to determine their common features.

Pharmacophore modeling explains structure activity relationships of a series of active molecules which forms a basis for the design of newer analogs.


Catalyst18 allows the manual definition of a pharmacophore using a single compound. Catalyst used to identify possible binding features between a receptor and a set of ligands that can explain variations in their activity.

Specific Features of Catalyst: *Specifies hypotheses in terms of chemical features that are likely to be important for binding to the active site. Each hypothesis consists of four parts: includes hydrophobes, charged/ionizable groups, and hydrogen bond donors/acceptors.

*Checks surface accessibility, so that it only considers those hydrophobic groups or hydrogen bond donors and acceptors that are available for interaction with the receptor.

*Defines the position of different features by absolute coordinates rather than by inter-feature distances alone.

Accelry's Catalyst18 can generate two types of Hip Hop and HypoGen depending on biological activities.

Hip Hop

Feature-based alignment of a collection of compounds without considering activity was provided by Hip Hop and it also attempts to derive a pharmacophore based on features that are common to active molecules19. Hip Hop takes a collection of conformational models of molecules and a selection of chemical features resulting in a series of molecular alignments in a variety of standard file formats. Hip Hop also maps partial features of molecules in the alignment set which gives the option to use partial mapping during the alignment. Partial mapping allows you to identify larger, more diverse, and more significant hypotheses and alignment models without the risk of missing compounds that do not map to all of the pharmacophore features. If a pharmacophore model is less likely to map the active compound, then it will be given higher rank; the reverse is also true20.


HypoGen based models have been successfully used to suggest new directions in lead generation - lead discovery and for searching a database to identify new structural classes of potential lead candidates. It creates SAR hypothesis models19-21 from a set of molecules for which activity values are known. HypoGen selects pharmacophore that are common among the active compounds but not among the inactive compounds .HypoGen generates hypotheses that are set of features in 3D space, each containing a certain tolerance and weight that fit to the features of the training set, and that correlate to the activity data.This algorithm generates an activity-based pharmacophore model which can be used to estimate activities of new compounds.

The hypotheses are created in three phases Constructive, subtractive and optimization phase19

Constructive phase

Constructive phase is very similar to Hip Hop algorithm. The constructive phase identifies hypotheses that are common among active compounds .All hypotheses (maximum 5 features) among the two most active compounds are identified and stored .

Subtractive phase

The subtractive phase removes hypotheses that are common among the inactive compounds by inspecting the hypotheses that were created in the constructive phase .

Optimization phase

The optimization phase attempts to improve the initial hypotheses . The optimization is done using the well-known algorithm simulated annealing.

Hypo Refine

Hypo Refine algorithm is an extension of the Catalyst HypoGen algorithm which is recently applied for generating SAR-based pharmacophore models .It also helps to improve the predictive models generated from a dataset by a better correlating hypothesis with the steric properties that contribute to biological activity. In addition, over-prediction of inactive compounds with pharmacophore features in common with other active compounds in the dataset can be overcome by HypoRefine .

Cost analysis

When an automated hypothesis generation was made to run, Catalyst considers and discards many thousands of models. It distinguishes between alternatives by applying a cost analysis. The overall assumption is based on Occam's razor; that is between equivalent alternatives, the simplest model is best. In general, if the difference is greater than 60 bits it represents a true correlation. Since most returned hypotheses are higher in cost than the fixed cost model, a difference between fixed cost and null cost of 70 or more is necessary to achieve the 60 bits difference19-20.

Types of Cost


- cost of the simplest possible hypothesis.

- fits the data perfectly.

Null Hypo cost

- cost when each molecule estimated as mean activity.

- acts like a hypothesis with no features.

Best pharmacophore model lies between the fixed cost and null hypo cost.Larger difference between pharmacophore and null cost corresponds to more significant models

Cost Components


- bits needed to describe the errors in the leads.

- the smaller the errors, the smaller the cost.


- bits required to describe the feature weights.

- the closer the average weight value is to expected typical value, the smaller the cost.


- bits required to describe the feature tolerance values.

- the closer the average tolerance value is to expected typical value, the smaller the cost.


- bits required to describe the types and relative positions of the features in the hypothesis.

- derived from pharmacophore space and possible combinations due to variable tolerance and weights.

Pharmacophore Validation

1. Percent yield of actives: % Y = Ha / Ht x 100.

2. Percent ratio of the activities in the hit list:% A = Ha / A x 100.

3. Enrichment (enhancement)

E = Ha / Ht = Ha x D

A / D Ht x A

False negatives: A - Ha and False positives: Ht - Ha

Goodness of fit = Ha (3A + HT) x 1 - Ht - Ha

4 Ht A D - A

Best hit list (i.e., Ht = Ha= A); False negatives = 0, false positives = 0.Worst list: Ha = 0, Ht = D-A) False negatives = A, false positives = D-A.


In modern computational chemistry, pharmacophores are used to define the essential features of one or more molecules with the same biological activity. A database of diverse chemical compounds can then be searched for more molecules which share the same features arranged in the same relative orientation

The Pharmacophore model has also been used in the below applications

Testing ideas (find new candidate molecule)

Prioritizing leads

Designing compound libraries

Predicting activity

Using the resulting alignments for additional studies

Search queries for database mining.