Xueli Liu, Ph.D., assistant professor in the Division of Biostatistics, has developed a new, “cinematic” way for researchers to see how drugs change gene activity. The method could someday help doctors refine and customize drug choices for patients and allow researchers to develop and improve new therapies.

The work, which Liu began in collaboration with Mark Yang, Ph.D., while at the University of Florida in Gainesville, Fla., appeared in the October edition of the journal Biostatistics.

Xueli Liu has created a way to progressively monitor gene expression. (Photo by Darrin S. Joy)

Cells turn genes on and off as part of their everyday life. They also ramp up and down gene activity — or expression — in response to different internal and external factors. Taking a drug, for example, can affect gene expression.

Understanding how a specific drug affects gene expression can help researchers better understand if and how the drug affects both healthy and diseased cells. It also could one day show physicians if a patient responds to the drug, allowing the physicians to change to a new therapy if necessary.

Scientists today use microarray technology to measure gene expression.

Microarrays consist of thousands of dots of genetic material arranged on a chip. Each dot holds a different genetic probe, and researchers expose the dots to tumor samples taken from patients. When a dot glows, it means the gene corresponding to that dot is active in the patient’s tumor.

Microarrays normally take a snapshot of gene activity, indicating which genes were active at the time the sample was taken.

Liu created a way to analyze gene expression that combines microarray snapshots from samples taken at different times after a drug is given. This collection of snapshots becomes a “video” of how genes change their activity over time after a patient takes a drug.

“By adding the dimension of time to how we analyze the data, we can give a more accurate picture of how different genes respond to a drug effect over the monitoring time period,” said Liu.

Creating gene expression videos requires tissue samples from many patients.

“We usually can only get a few samples from any one patient, so we have to combine many patients’ samples to get a complete view of gene expression,” explained Liu. In effect, Liu combines several low-quality gene expression videos to get one high-quality video.

However, once researchers have the high-quality combined video, they can turn around and use it to enhance each individual patient’s video and get a clearer picture of how the patient’s genes respond to the drug.

Because the new technique appears to outperform current gene expression analysis methods, Liu hopes it will speed researchers’ efforts to develop more effective treatments.