Models connect technology, biology
Tracy Barger, Cavalier Daily Senior Writer
As the new millennium begins, the computer and biological worlds are moving closer together.
By some time this century, new techniques of DNA-based computing may allow computers to laugh at jokes and recognize the difference between cats and dogs. Computer modeling of genetics will equip the simulation of systems with a huge number of variables, allowing scientists to better understand how genes work.
In a lecture last Thursday in the Rotunda Dome Room, Evelyn Fox Keller, history and philosophy of science professor at the Massachusetts Institute of Technology, spoke about the growing connections between genetics and computers.
"Computational models have the particular potential for yolking computers and organisms into ever more literal proximity," Keller said.
In recent years, computer models have provided scientists with a way of understanding the complex interactions in molecular biology. Studies of molecular biology, particularly genetics, involve large amounts of experimental data and the use of computational computer models can be beneficial in attempting to interpret this data, Keller said.
Biological computer models have the dual purpose of exhibiting the results of previous experiments and suggesting future experiments, resulting in a constant back-and-forth between the process of experimentation and model development.
Computers are also useful in studying genetics because the genetic regulation process operates in a manner similar to computers. Both work as logic circuits with actions being determined by a multitude of true-false type variables.
A computer model of genetics predicts what processes a gene will trigger based on the presence of various gene regulatory compounds.
Despite these similarities and potential benefits, there are restrictions to the use of computer modeling in biology.
"Computers like numbers to be the same from day to day; cells are not really like that," William Pearson, prof. of biochemistry and computer science said.
Despite these differences, the similarity between the processes of gene regulation and computer operations could lead to applications in both molecular biology and computer science.
"A number of computer scientists have begun to look to molecular biology for new horizons in computer technology," Keller said.
One area where molecular biology has become beneficial to the development of computers is in the new field of DNA computing.
In this field, scientists use pieces of DNA in a test tube to solve complex problems.
Scientists synthesize the DNA in a test tube and then use this genetic material to encode certain computations, said Gabriel Robins, University assoc. professor of computer science.
"DNA computers exploit the chemistry of DNA," he said.
DNA computing has benefits over the standard computer in tasks that involve massive parallel processing. It is not as efficient as a conventional computer in some tasks such as adding all the numbers in a spreadsheet, something a conventional computer could do in a few seconds, he added.
Robins said he envisions scientists using DNA computing only in special situations rather than replacing conventional computers altogether.
"The best case scenario is that it is a very specialized technology," he added. "Maybe in a hundred years" there will be "a biological computer peripheral augmenting the conventional computer."
This type of technology could have advantages in performing tasks that do not have well-defined data, such as understanding a joke or discerning the difference between an image of a cat and dog - tasks that historically have been very difficult to program on conventional computers, Robbins said.