Bioinformatics-Virtual Drug Development

Bioinformatics - Drug discovery   Bioinformatics- Sequence analysis 

Bioinformatics-Virtual Drug Development

These days, computers are an integral part of genomics-based drug discovery, helping researchers find drug targets by comparing databases of genomic information with annotations about functional information, by analyzing the data that comes in from various wetlab experiments, and by simply keeping track of the huge amounts of biological data being unearthed in life sciences research. This is the role of bioinformatics, a field that has exploded in importance over the last few years as companies have begun to realize they are drowning in raw data.

But now the uses of computers for other parts of the discovery and development process are coming to the fore. Theoretically, researchers could now test virtual drug compounds against virtual protein targets, study the virtual pharmacokinetics of their optimized virtual lead in what amounts to virtual animals, study its effects on virtual organs, design a virtual clinical trial to test assumptions and variances, and even answer some regulatory questions through simulation. Somewhere in that process, a chemist has to actually mix up a compound and conduct some experiment but buckets of silicon are being added to the discovery and development process every day, with the hope that the wet lab will one day become as dry as a sand box.

Twentieth century biology has been about cataloging the elements of life. Every day, we have a little more of the recipe of life, stretching before us as an almost endless line of As, Gs, Cs, and Ts--forming general sequences common to most living organisms, gene sequences common to most humans, polymorphisms peculiar to small subpopulations. But this static information amounts to little more than a parts catalog, a shopping list for a living organism. A vital thrust of 21st century biology will be the animation of these static parts.

After all, a long string of base pair letters is like well a long string of letters. It makes for a less interesting read than a telephone directory, and while it tells you how dial up all sorts of important proteins, most sequences alone tell you little more about a person than does their phone number. We cannot yet predict protein folding from amino acid sequence, nor can we accurately predict protein function from protein shape. We can, of course, correlate certain polymorphisms with likely disease outcomes, and we learn more every day. But the more we learn about the importance of these new variables, the more we have to take into consideration when developing clinical strategies, undertaking drug development, and designing clinical trials. And gene sequence, even when linked to functional information, will only be one of many variables to consider in optimally designing therapeutic interventions and treating disease.

One way of animating our growing store of static information is through computer simulation. It is an area that is beginning to emerge slowly in the life sciences, with only a handful of academic and commercial players active in the area. But for a fledging discipline, there is a great variety in the scope of work being undertaken. While academic labs try to create accurate simulations of red blood cells and simple bacteria, the private companies are taking on bolder projects--simulating human organs and even human diseases in their entirety.