Introduction to genefinding

Just under fifty years ago two discoveries were made that have changed, or will soon change, the way the human race understands and experiences life. In 1947 Bardeen, Brittan, and Shockley discovered the transistor effect, an achievement for which, in 1956, they received the Nobel prize. A few short years later, Jack Kilby at Texas Instruments (1958) and Robert Noyce at Fairchild Camera (1959) made the breakthroughs from which the integrated circuit (multiple transistors on one substrate) was later developed.

In 1953 Watson and Crick unlocked the structure of the DNA molecule and set into motion the modern study of genetics. This advance allowed our study of life to transcend the wet and dirty realm of proteins, cells, organelles, ions, and lipids, and move up into more abstract methods of analysis. By discovering the basic structure of DNA we had received our first glance into the information-based realm locked inside the genetic code.

For almost fifty years genetics and computer science coexisted side-by-side without much interaction. Computer science is a discipline of abstract data, concerned with its manipulation, management and analysis. Genetics was empirical in nature and experiment driven; analyzing large amounts of data was simply not a problem geneticists had to contend with.

Historically, for geneticists, the amount of time required to generate data has been much greater than the amount of time required to analyze the resultant data. Genetics experiments that necessitate weeks, months, or even years to bring to successful fruition can generate as little as a few radiographic films in data. After a successful experiment, a triumphant geneticist would leisurely (between writing grants and teaching classes) scour the data as a hungry dog would attack a leftover steak, analyzing every available nook and cranny, leaving no promising datum untouched. In the past, generating good experimental data was analogous to searching for gold in that it was scarce and highly-appreciated if it was found.

One problem geneticists are not accustomed to contending with is an over-abundance of information. In some ways, the reluctant pace of incoming information has been a good thing for genetics; the interpretation of genetics data can be a challenging endeavor and the relatively slow tempo of its acquisition has allowed ample time for researchers to fully appreciate the implications of each small piece as it arrived.

Human beings and computers have divergent and complimentary abilities. Computers are intrinsically beasts of information; they deal with pure abstract data, ones and zeros. Relative to humans, computers excel at manipulating large amounts of data, performing numerous calculations quickly, and analyzing large, multi-dimensional data sets. Humans, by contrast, are physically rooted in nature and have a proclivity for higher abstract thinking, long-term planning, and assimilating noisy or incomplete information. We are flexible and adaptable where computers are efficient and rigid.