The field of bioinformatics can be divided into "bread and butter" and "blue skies" applications. The bread-and-butter material is the application of computer-science techniques to the organisation, storage, retrieval and searching of mountains of bio-data. Throw intuitive user interfaces and data visualisation into the mix and you've pretty much covered what 80 per cent of the current software does.
But a new breed of software is coming to the life sciences that will help to revolutionise the way research is performed and commercialised. I'm talking about the computational modelling of complex biological processes: so-called Big Biology.
If a researcher has a promising new drug with an identified drug target, for example, and wants to test it, how will they do it? Animals provide the best-known surrogate for humans in the laboratory. At a broad level, all mammals (including humans) are extremely similar, reflecting our common origin. This is true of circulatory systems, locomotory and sensory systems, and so on.
This approach is flawed on many important fronts: there are animal-rights issues to consider; drugs must, in any case, be tested eventually on humans; and drug companies do not save money by using animal models, as the cost of running the programs is extremely high.
So what's the alternative? Looking into the crystal ball, eventually software will be sophisticated enough to model complete organisms virtually - in software. The only variable is time: will it take 10, 20 or 40 years? Who knows? But it will happen.
The forerunners of these software systems already exist. Research groups in Europe, the US and Japan are working on modelling metabolic pathways (see glossary) in software. Eventually these systems will scale up to model multiple, interconnected pathways and processes, as a realistic model needs to show how multiple pathways interconnect and affect each other.
What are the stumbling blocks to building these advanced models? As I see it, there are two main problems that need to be overcome.
The first is a lack of understanding of the "business" of biology. There are some incredible statistics. We still don't know what 40 per cent of proteins in our body do in isolation, for example, never mind putting them together in a huge system of almost unbelievable complexity.
Nuclear receptors are not fully understood, yet they represent perhaps the single most important category of drug targets in the human body. These are just some examples. The bottom line is that you can't model anything in software until you understand how it works in the real world.
And then there is the inability of existing technology to perform tasks required of it adequately. Current computerscience techniques simply will not scale to the trillions of computations required per second to model this kind of software. Forget the latest Pentium processor or Nintendo box: we're talking about petaflops (see glossary).
Existing hardware and software methodologies will probably be unable to meet the requirements. Other technologies from physics and chemistry will be needed. Two good examples are quantum computing and nanocomputers (see glossary).
CUTTING-EDGE techniques such as these - and other technologies not even dreamed about - will be required to move the idea of accurate virtual organisms from fiction to reality.
Eventually, though, I believe a researcher will be able to run thousands of test on virtual organisms in parallel, perhaps with each experiment looking at a different potential side effect of a new drug or testing adverse reactions to different dosage levels. The list of possibilities is endless.
Now consider this: if software is powerful enough to faithfully emulate, in every detail, a living organism, then does that virtual organism have the same rights as a real organism? It sounds like a stupid question, but what if the virtual organism were a human that consistently showed emergent and intelligent behaviour? The ramifications would be endless, on scientific, theological and ethical levels.
Moving away from the advanced-modelling aspects and back to where you will see the impact of biotechnology and bioinformatics on a more individual note, your doctor and hospital will be increasingly connected to a decision-support system designed to help them to make the right decisions for your health.
Whether it's to get a second opinion on your clinical symptoms or to search an online database to check your genetic susceptibility to a disease, technology that today is regarded as cutting-edge research will be commonplace in everyday health-care procedures.
SO WHY is all this exciting? Bioinformatics is exciting at every level. The computing challenges are enormous. There are chunks of knowledge missing - and doesn't not knowing the answers make a refreshing change from developing boring websites?
It's also impossible not to get involved in the area: researchers are beginning to unravel the essence of life and humanity. Whether or not you believe in God, it feels weird to look at the information and structures encoded in a DNA sequence and see that some of the error-checking techniques and encoding paradigms bear a striking resemblance to methods that are commonplace in computing. It makes you think.
Biological modelling in software is as hard as it gets. Forget the Internet gurus. In bioinformatics, using Internet technologies is a given - it's the fundamental architecture you use to build systems, and it rarely gets a mention. Top "bionauts" are as comfortable working with databases, Web servers and Unix as they are reading extracts from Nature or New Scientist.
There are no two ways about it. There are questions to be answered about harnessing the power of biotechnology in an ethical way - and computer science will get sucked into the argument soon. There are two ways to look at it. Either the doomsday scenario that some predict will come to pass - "geno-racism", for example - or biotechnology will fold seamlessly into our daily lives. As always, the truth lies somewhere between the two extremes.
Biotechnology and bioinformatics will have a fundamental impact on how we live our lives in the future, dwarfing the impact of the Internet. No one knows the answers to the questions that lie ahead. More to the point, many of the questions aren't yet known. There is one thing we can be sure of, however: computer science will play a key role in the quest to understand ourselves.
Humphrey Sheil is the chief executive of Teogas Systems (www.teogas.com), a bioinformatics company based in Dublin.