WASHINGTON—Using a novel data mining algorithm they call Pattern Array software, researchers from the University of Maryland's School of Medicine are studying movements in rats that may help them predict diseases such as amyotrophic lateral sclerosis (ALS) and test therapies aimed at preventing, slowing or stopping disease, starting with hereditary forms.
Publishing their findings recently in the American Psychological Association journal Behavioral Neuroscience and the Nature Publishing Group journal Neuropsychopharmacology, a team of researchers from the Maryland Psychiatric Research Center says because the original software can be therapeutically useful even before science understands how disease begins, they may seek a commercial partner to bring Pattern Array to market.
"We've been developing this over the course of several years, but it's been a matter of convincing ourselves, manuscript reviewers and grant reviewers that it truly works," says Dr. Greg Elmer, a lead author of the studies. "We really do think it's pretty novel, and would serve well a pharmaceutical company that wants to discover novel compounds that have therapeutic value that you wouldn't normally suspect from its structure."
Mathematically analyzing about 50,000 predetermined movement patterns that resulted when rats and mice roamed freely in a small arena, the software used by the researchers created an abstract space defined by combinations of behavior such as speed, acceleration and direction of movement. Mining the resulting behavioral data enabled researchers to test many more facets of behavior than they could analyze manually.
In the Behavioral Neuroscience article, the authors demonstrated their original software on mutant rats used as an animal model of ALS, the progressive and fatal neurodegenerative condition known as Lou Gehrig's disease. After videotaping the movement of two groups of rats—one with the mutation that results in an ALS-type syndrome, the other normal controls—the scientists used the computer to "pan" for differences between groups and identified a unique motor pattern in mutant rats two months before disease onset, which would equate to roughly five to 10 years in humans.
"This is significant because the problem with ALS is despite having symptoms of the disease, it usually takes a long time before the disease actually appears," says Dr. Neri Kafkafi, another lead author of the studies. "Being able to predict more accurately which carriers may express the disease before they experience symptoms, researchers could test medicines that might prevent symptoms from emerging," he says.
In the Neuropsychopharmacology article, the authors showed that Pattern Array can also be applied for mining behaviors predicting additional properties, such as within-class differences between drugs and within-drug dose–response, all of which can be measured automatically in a single session per animal in an open-field arena. This suggests a high potential for Pattern Array to be effective as a tool in psychotherapeutic drug discovery, Elmer says.
"We believe the most interesting aspect of Pattern Array was that it classified drugs into three psychopharmacological classes—psychomotor stimulants, opioids and psychotomimetics—using a single behavioral assay, while in 'traditional' drug discovery, that would take at least three different assays, or one for each class," Elmer says. "Once we build our database additional drug classes such as antipychotics, antidepressants, anxiolytics, et cetera, can be predicted within the same assay."
Elmer points out that in its current form, the scientists consider Pattern Array to be a discovery strategy more than a software package. The research team has immediate plans to continue to improve the Pattern Array algorithm and broaden the spectrum of therapeutic entities in the database and will begin to consider commercial partners to develop the software.
"We are a bit of an anomaly, so with the help of our Office of Research and Development, we are breaking trail," Elmer says. DDN
