Stanford Research Communication Program
  Home   Researchers Professionals  About
Archive by Major Area


Social Science

Natural Science

Archive by Year

Fall 1999 - Spring 2000

Fall 2000 - Summer 2001

Fall 2001 - Spring 2002

Fall 2002 - Summer 2003




I-RITE Statement Archive
About I-RITE

Looking for "Switches" for Genes

Yueyi Liu
Department of Biological Sciences
Stanford University
March 2002

In our bodies, whether genes make their products (usually proteins) is tightly controlled by "switches", regions that are capable of turning genes on or off. Malfunction of these "switches", so called regulatory regions, often lead to serious diseases such as cancer. My research is to identify these regulatory regions based on the fact that these regions tend to be very similar in different species, such as human and mouse. This approach is different from the traditional experimental biology approach because it is based on computation. It has the potential of identifying regulatory regions on a large scale.

In our genome (the entire DNA sequence in our body), there are short sequences of DNA that regulate whether genes make their products. These short sequences, or regulatory regions as they are known in the scientific community, are of crucial importance for our bodies to function normally. Many diseases are caused by the malfunction of these regions. One such disease is Burkitt's lymphoma, a form of cancer that primarily affects small children. Without treatment, it often leads to death. It occurs when a specialized white blood cell becomes cancerous. This is caused by a disruption in the regulatory region that controls the making of the myc protein. The myc protein is normally made only when there is a need for cells to divide. But in patients with Burkitss's lymphoma, the myc protein is constantly being made, causing cells to divide uncontrollably. This uncontrolled cell division eventually leads to cancer. It would be great if we could pinpoint the location of the regulatory region, and develop drugs that target it to restore its function.

However, these regions are notoriously difficult to find. In the past, the only way biologists could find them is by first locating the gene of interest, then modifying the regions surrounding it randomly until no proteins are made from this gene. As you can imagine, this is a very laborious process.

Fortunately, biologists have noticed that regulatory regions, like the genes they control, tend to remain similar in different species. For example, the human hemoglobin gene and the mouse hemoglobin gene are 95% identical in sequence. Their respective regulatory regions are very similar too. This knowledge, combined with the recently completed human genome and several other genomes, provide us with a powerful tool to find regulatory regions using computational methods.

My research focuses on finding regulatory regions using computational methods. I make use of available computational programs to look for regions that are of high similarity in two or more species (e.g., between human and mouse). These similar regions are often potential regulatory regions. I will then collaborate with biologists to verify by experimental methods whether they are indeed regulatory regions. With this method, thousands of regulatory regions can be discovered at the same time, rather than one by one.

However, there are two problems associated with this relatively simple approach. The first is that some regulatory regions may be missed, due to low similarity between the species compared. The second is that I may find too many regions of high similarity in certain parts of the genome, many of which are extraneous. The huge number of potential regulatory regions that needs to be verified experimentally could potentially delay the discovery of true regulatory regions.

To address the first problem, I am currently doing a comprehensive study on all the regulatory regions discovered by experimental methods. For each regulatory region, I will check to see if it is similar in several species. This study will give me a clear idea about the kinds of regulatory regions that are not similar, establishing the limit of my method. Furthermore, this will give insight on new methods that need to be developed for discovering these kinds of regulatory regions.

To address the second problem, i.e., too many potential hits to verify, I will again turn to the experimentally determined regions. By taking a closer look at these regions, I will have a better idea of what regulatory regions should look like. This will help me weed out those regions that, though similar, are unlikely to be regulatory regions.

I am confident that my research will lead to the speedy discovery of many regulatory regions. It will not only help contribute to our understanding of some basic biological processes, but also help identify the cause for many diseases, therefore helping to find the cures for them.