dynamics of cell behavior. For example, a certain protein P (produced by a gene G) is used to prevent the production of another protein P ’ by blocking the processing of gene G ’. The computer science analog is keeping some parts of a program from being executed once the execution reaches a certain stage. This behavior is akin to Edsger Dijkstra’s semaphores in regulating the dynamic behavior of concurrent programs [ 2].

Both systems and synthetic biology will require expertise in computer science way beyond the expertise necessary for processing sequences. In turn, these disciplines will challenge computer scientists with

ing is measured by special scanners linked to computers. The measurements estimate the amounts of products generated by a gene. Microarrays can also be used by biologists to dynamically record the changes in gene products over time, as a cell is subjected to some external influence (such as food starvation or the effect of a drug).

Microarrays are also expensive, each costing hundreds of dollars, and tens or possibly hundreds of them may be necessary to study the gene interactions of a single cell. But, as with sequencing costs, microarray costs are decreasing, and the volume of data being generated by microarray experiments is gigantic, possi-

 

The computer algorithms in systems and synthetic biology are akin to those

used for finding bugs or incorrect behavior in large complex programs.

 

problems that are at the forefront of computer science today, including how to develop nanotechnology hardware, fault-tolerant circuit design, program verification, model checking, program synthesis from data, and data mining.

Systems biology deals with gene interactions, some involving hundreds, if not thousands, of genes. When some interactions go awry, cell behavior changes dramatically, resulting in situations like the uncontrollable growth of a cancer. Thus, the computer algorithms in systems and synthetic biology are akin to those used for finding bugs or incorrect behavior in large complex programs. Debugging is one of the most arduous tasks in program development. Nonetheless, computer scientists have already developed sophisticated tools to facilitate debugging, some applicable for finding faulty configurations in biological networks.

A microarray, or genome chip, is a silicon chip that has become a common tool in systems biology. It includes tens of thousands of minute wells, each with multiple short strands of DNA material representative of each gene ( www.affymetrix.com/index. affx). Each strand is matched with its counterparts obtained through a wet-lab experiment involving the genes of the cell being studied. The degree of match-

bly surpassing the size of the available DNA data. Even though results of microarray experiments are coarse and may contain laboratory errors, they represent challenging problems for data-mining experts.

Several research groups led by prominent computer scientists have immersed themselves in fascinating research involving the amalgamation of computer science and systems biology. For example, a team at the Weizmann Institute led by Ehud Shapiro has designed nanobiological processors that function as finite-state-automata to recognize desired sequences of DNA and eventually deliver drugs capable of correcting cell behavior that could lead to disease ( www.wisdom.weizmann.ac.il/math/profile/ scientists/ shapiro-profile.html) [ 8].

As mentioned earlier, synthetic biology is a notable leading-edge effort in systems biology. Its aim is to use the existing “processing” capabilities of a cell (such as yeast or E. coli) to perform such tasks as cleaning the environment, detecting dangerous chemicals, and manufacturing drugs. J. Craig Venter, a pioneer in sequencing the human genome, is pursuing the goal of generating (in a wet-lab) very long sequences of artificially produced DNA (www.jcvi. org). The artificial DNA is being designed to perform the tasks involved in reengineering cell behav-