While biocomputing has an old and respected history in structural molecular biology, the recent boom is mainly a consequence of the large sequencing projects, like the Human Genome Project. These projects yield an enormous amount of DNA and protein sequence data now available from public databanks.
This amount of data is shifting research in molecular biology and genetics from a purely experimental approach to one in which experiments can be planned in front of a computer. The pace of biological research speeds up, creating new economic opportunities for companies, and new professional profiles in industry and academia. In addition, private sequence databases are beginning to appear, often containing more information than the original public ones. This is a clear sign of the growing economic importance of biocomputing.
In particular, one widely unknown kind of privately held data are so-called Expressed Sequence Tags. Instead of sequencing a whole genome or a part of it, the idea is to sequence bits of DNA which represent genes expressed in particular cells, organs or tissues of different organisms. They yield the "dynamic" picture of the gene expression patterns, as opposed to the "static" picture emerging from genome studies. Big biotech companies like Merck and Smith, Kline & Beecham have recognized the enormous potential of the information coming from this new technique. The total number of Expressed Sequence Tags in private databases is approaching 1.000.000, two times the size of what is publicly available, and new computational methods and tools are being developed by biocomputing companies to exploit this information.
In the context of economics, this essay will examine the importance of biocomputing for medicine and agriculture, and the impact of biocomputing on science in general.
Biocomputing techniques offer opportunities to biotech companies which want to give a stronger rational basis to the process of drug discovery. Experiments can be designed much more intelligently and the understanding of molecular interactions can be enhanced dramatically. Trial-and-error experiments can be replaced by predictions that allow for the design of custom-made drugs. The development of safer, 'smarter' and 'greener' drugs is the result. (For an introduction to rational drug design, please take a look at Wolfram Altenhofen's article in this volume.)
Drugs based on rational design are 'smarter' in the sense that they are optimized for the target molecule in question and therefore tend to elicit far fewer side effects, which is a major problem for many therapies (e.g. cancer and AIDS therapies). Smarter design also calls for lower doses of drugs to achieve the desired effect.
Many custom-made drugs may also be called 'greener' because their industrial production can be achieved in a way that is compatible with the demand for environmentally clean and safe production methods. With a greater understanding of the processes that govern the synthesis of biomolecules, we are able to use a lower amount of detergents, solvents, and energy to obtain the same quantity of product.
Furthermore, access to public databanks/resources improves the exchange of information on drug design research, and promotes cooperation and reduce waste due to unnecessary competition, duplication and re-inventing the wheel.
Agriculture can exploit the same computational tools, databases and strategies for investigating disease resistance markers, pest resistance genes and so on. As a result, crop harvests can increase and special plants can be engineered to alleviate food shortages in developing countries.
Because of all these indirect implications, the current economic cost of biocomputing will seem almost irrelevant, e.g. to the contribution on both the "health of the nation" and hopefully the nation's health bill. In agriculture, the same return on investment can be assumed.
For example, biocomputing research challenges many of the current pattern recognition techniques. The difficulty of current problems in bioinformatics provides a significant testbed for developing techniques for machine learning and the recognition of complicated patterns.
The biologists' need to retrieve large amounts of data from different datasets and to correlate them is pushing ahead current database technologies: the development of one of the first object-oriented databases comes from biology, and the same holds for the first integrated databases accessible from the WWW.
Of course, all these developments eventually create a significant economic impact.
Biocomputing, like any form of basic scientific research, has both the immediate potential to produce economically important spin-offs and the potential to have far-reaching positive effects on world society and markets.
In the short term, the immediate application areas described above yield a significant, though mostly indirect impact on the economy. In the long run, biocomputing is an excellent candidate for a good facilitator to re-vitalize the pharmaceutical, chemical and agricultural industries, to attract new talents and ideas to computer science and biology, and to produce a positive fallout of ideas and new technologies to society in general.
The authors would like to thank everyone who has contributed ideas and shared opinions on this essay, in particular to Wolfram Altenhofen, Nikolaj Blom, Paul Brennan, Lew Gramer, Peter Hjelmstrom, Lau Chin Hoon, Jörn Kalinowski, Michael Lappe, Rebecca Parsons, Karsten Quast and Alexander Sczyrba.