IT in Pharma - Special

Mining a molecule

Drug discovery need not be based on serendipity. Information Technology has made it possible for researchers to model, analyse and shortlist potential new drug candidates. Katya Naidu assesses the various bioinformatics solutions, which have become an integral part of pharma R&D.

Computer aided drug design and bioinformatics have become an integral part of the drug discovery process. Many pharma companies have understood the importance of technology-based assistance to intuition which governs drug discovery.

"Bioinformatics has crossed the chasm and is today, a mainstream technology. They are well integrated into the discovery process," states Sowmyanarayan, Senior Manager-Business Development and Alliances, Strand Lifesciences.

Aiding research

The utility of bioinformatics solutions kicks off at the target identification process and spreads its services throughout numerous areas in drug development like, target validation, study of molecular basis of diseases and rational models for drug design. "Sequence searches, profiling of targets, computations on large and varied signaling pathways to identify altered pathways due to differential gene expression would be the areas, where bioinformatics are useful in target selection and validation," says Sridhar Mosur, CEO of Jubilant Biosys.

IT is also useful in areas of lead optimisation in the form of chemistry databases. These databases have information on the chemical structure of compounds and their interaction with targets, which are of high value to the process of drug discovery. If it is known that a drug must bind to a particular spot on a particular protein or nucleotide, then a drug can be tailor-made to bind at that site. This is often modelled computationally using any of a number of techniques. "These techniques attempt to reproduce the researchers' understanding of how to choose likely compounds built into a software package that is capable of modelling a very large number of compounds in an automated way," says Anuradha Acharya, CEO, Ocimum Biosolutions.

Once a number of lead compounds are found, computational techniques help refine the molecular structures to choose molecules with greater drug activity and fewer side-effects. A computational technique called Quantitative Structure Activity Relationships (QSAR) is useful in detecting the functional group in a compound. "This can be done using QSAR, which consists of computing every possible number that can describe a molecule and then doing an enormous curve fit to find out which aspects of the molecule correlate well with the drug activity or side-effect severity," says Acharya. This information can then be used to suggest new chemical modifications for synthesis and testing. "Ideally, there is a continual exchange of information between the researchers doing QSAR studies, synthesis and testing," Acharya adds.

Expression analysis is another key process in drug discovery. Microarray data analysis packages for statistical and biological analysis ease the process of target identification. "Functional classification, pathway analysis and annotation supplements aid researchers in deriving biologically significant knowledge from expression data. Expression data, when mapped onto these pathways gives further insight into pathways and steps influenced by affected genes. "Studying pathways in relation to expression data helps in identification of possible drug targets as they provide an insight into how rate-limiting steps are being effected," says Acharya.

Modelling drugs

Modelling of predicted pathways is yet another solution in getting closer to the discovery of the right candidate, that could make it as a drug. "Metabolic, proteomic, genetic, cellular, and pathway events are all interdependent in an organism and they are in a flux. Therefore, modelling these events is important. Biochemical pathway and systems analysis for target identification and validation is an area that would help drug discovery and development extensively," asserts Mosur. This exercise examines various types of biochemical pathways, including metabolic and signalling pathways and gene regulatory networks. Structural homology modelling along with pathway information helps in identifying the mechanism of drug action.

Mathematical models, which are constructed based on dynamic changes in metabolite concentrations over time and stages, enable the prediction of pathways. Simulation modelling helps in analysing reaction kinetics of pathways. Simulations explore multiple hypotheses of disease progression and treatment. "The idea behind simulations is to use computers to understand the biology behind a disease and then correlate the biology to drug response," says Sowmyanarayan. In every simulation that is performed, one can not only understand how the overall system (human) behaves in the disease but also exquisite details on how every aspect of the biology is affected by the disease and its response to a drug.

Speaking genetics

Progression in genetics and gene-based drugs has aided the need and popularity of bioinformatics. Understanding the human genome and genome of several other organisms has led to a stronger area of genomic research. Understanding the genome in depth, due to the vast information, is not possible without the use of computers. Hence, computers and bioinformatics play a significant role in genetic research," says Mosur.

Bioinformatics comes to the rescue of highly complex high-end genetic research. One such example is in the discovery of drugs based on RNA interference (RNAi). This intracellular gene level process involves the prediction of the pathways of siRNA, which is minuter version of RNA.

The current focus of the research is on the design of arrays that generate silencing profiles for pathways, diseases and metabolism.

Although the success of the drug depends on identification, most organisations can design not more than three siRNAs per gene to make experimentation practical and economical. There are hundreds of targets identified for any typical mRNA. The difficulty in screening for such few drug candidates from the hundreds of targets identified for any typical mRNA has necessitated recognition of features that weed out ineffective siRNAs. Ocimum Biosolutions has come up with a solution called iRNAchek that provides a way to design test and analyse siRNAs for success criteria.

Discovering right

Bioinformatics solutions give an entirely new dimension to the field of research. Studies done using bioinformatics, are speedy as they rely on computational power.

They help reduce the overall cost of drug discovery by enabling researchers to find out sooner that their 'hoped for' compound is not working out and steer them towards more promising candidates. "Applying bioinformatics solutions could cut down drug discovery process by at least two to three years and also allow scientists to work around large number of targets as well as predict failures," says Mosur.

Boston Consulting believes that bioinformatics can cut $150 million from the cost of developing a new drug and a year off the time taken to bring it to market. "R&D investment in pharma industry is acknowledged as high at levels of $800 million for a typical lead to drug conversion. Bioinformatics was embraced as a weeder that reduces time and effort spent in screening of leads," says Acharya.

editorial@expresspharmaonline.com