The biggest advantage of combinatorial chemistry over classical synthetic chemistry is that it can lead to compounds that otherwise might not be synthesised using traditional methods of medicinal chemistry.

Creating a large library of chemicals with novel structures is called molecular diversity (MD). Molecular diversity thus takes advantage of:

• the ability to isolate target molecules in pure, crude extract or whole cell in vitro assay screens, and
• development of robotics and instrumentation to perform high-capacity screening on microtitre plates in a rapid and automated fashion.

Today pharmaceutical companies have started using MD as an extension of their traditional work. Biotechnology companies use MD techniques to progress from molecular biology of large molecules to small molecules. The basic strategy of MD involves the synthesis of large compounds libraries from peptides, oligonucleotides, carbohydrates, to synthetic organic molecules.

There are three strategies, which are used for generating molecular diversity. All methodologies assemble every possible combination of given set of molecular building blocks, simultaneously recode those that have been used and then assay the resulting compounds simultaneously and select from the record, those, which are promising.

• Those using mutatable molecules (peptides and oligonucleotides) and the process of directed molecular evolution to rapidly optimise promising molecules that can act as templates for next round of optimisation.
• Those using small organic molecules as building blocks, which can not be mutated but exhibit large range of properties.
• To create and refine large numbers of peptides or oligonucleotides until a nanomolar-range molecule is found and then to convert that 'lead' into small organic compound by using drug design methodologies.

Two synthesis technologies allow rapid creation of combinatorial libraries. These are:

• 'Mix and hit' synthesis, and
• 'Parallel' synthesis.

Identification of active molecule is done by use of monoclonal antibodies or with use of soluble receptor. Fluorescence tagged other antibody can be used to locate earlier attached monoclonal antibody, using microscopy.

Computer-assisted drug design

The role of computational drug design is to aid in the discovery and optimisation of new candidate drug molecules.

The drug discovery cycle can be split into six phases:

1. Discovery and lead generation (1-2 years)

2. Lead optimization (1-2 years)

3. In vitro and in vivo assays (1-2 years)

4. Toxicology trials (1-2 years)

5. Human safety trials (1 year)

6. Human efficacy trials (1-2 years)

Thus total 6-12 years are required for the development of a drug and costs are $100-200 mn or more. Computational drug designing will mainly contribute to improve upon this cycle. Computer-based methods used for drug design may be for engineering of proteins, peptides, oligonucleotides or small organic molecules.

Computational drug design includes:

• New compound discovery by computer searching of chemical databases;
• Quantitative modelling of chemical behaviour of compounds - search for generation of structures and their conversion to 2- and 3- dimensional structures; analysis of activities or properties; tools for analyzing experimental data (such as spectroscopic or diffraction studies); modelling and visualization systems for examining and predicting chemical properties and structures;
• Compound optimisation by systematic modification of functional groups to maximise potency and to minimize or eliminate side effects and toxicity, and
• Computer-assisted de novo drug design (generation of entirely new molecules that might fit a receptor site and act as antagonists or inhibitors).

Most companies use computers in some part of their drug discovery and drug optimisation process.

The US company Net Genics has secured $6.5 million of investment to help it develop software that can be used to generate new drugs from genetic data. Net Genics will speed up the work on Synergy, its cross-platform bioinformatics software.

NaviCyte Inc. has entered into an agreement with SmithKline Beecham plc (SB) to collaborate on use of NaviCyte's proprietary computational biology tools to identify new candidates for drug development. One of the major problems in drug discovery and lead compounds selection is the high level of uncertainty in predicting the absorption and availability of drug in humans by using animal models. NaviCyte's computational tools are designed to increase the predictive power of animal models by improving the selection process of lead drugs chosen for animal testing in the first place. By using data obtained with proprietary in vitro High Throughput Pharmacokinetic technology in concert with computational tools, it provides valuable absorption information normally available after expensive animal testing. Also it comes at very early stage in drug discovery process. The result is that once the lead compounds enter expensive animal testing they are more likely going to correlate favorably with subsequent human trials.