Scientists have found a better way to study Ebola in the lab. When they infected a fleet of specially bred mice with the virus, they saw a range of reactions, from zero symtoms to death by hemorrhagic fever. By studying how different genes in mice change the course of infection, they could determine what genes make humans more vulnerable to death by Ebola.

The success of these new mice are described in a study published Thursday in Science, which presents the first truly viable mouse model for studying Ebola. In developing treatments and vaccines for a virus -- or even just trying to understand how it works -- researchers need to infect animals with it. Until now, the best animal models for Ebola have been monkeys.

But monkeys are large, expensive to keep, and more dangerous to work with than tiny rodents. And they also present ethical issues.

"Ask any scientist who does this type of work, and you'll hear that they really want non-human primates to be a last resort," said study co-head Angela Rasmussen, of the University of Washington department of microbiology.

But traditional lab mice just don't work: Researchers have been able to modify Ebola so that it infects and kills them, but they don't exhibit any of the traditional symptoms of hemorrhagic fever before they die. With the disease acting so differently from the human version, it's hard to use the mice for research.

And while guinea pigs and hamsters have been made to show symptoms, their genomes are much less intricately studied than mice. Scientists know the mice genome in its entirety, so it's easy for them to pinpoint what genes are involved in the disease and its treatment. Not so with these other animals.

"Unfortunately, monkeys were just the gold standard," Rasmussen said.

Instead of standard lab mice, Rasmussen and her colleagues used a group of genetically diverse mice created by the University of North Carolina at Chapel Hill. These mice were specially bred with disease-related testing in mind. While most lab mice only represent about 10 percent of the genes in the total mouse population, this group of mice cover 90 percent of the species' genetic diversity.

That means that testing viruses on them is something like testing viruses on the human population: The genetics of the host vary, and so does the outcome of the infection.

Sure enough, the mice reacted differently to their exposure to Ebola -- and for the first time, some of them (40 percent) showed the traditional symptoms of the disease, including death by hemorrhagic fever. Another 40 percent died without these symptoms, as regular lab mice would, and 20 percent were resistant to the disease.

In addition to encouraging other Ebola researchers to work with these mouse models, Rasmussen and her colleagues are beginning to investigate just which genes make a mouse more or less likely to become ill with the disease. Some of these variations may be applicable to humans, or at least point researchers in the right direction.

"This paper isn't earth shattering, but it's the first step in being able to do this kind of genetic analysis in humans," co-study head Michael Katze said. "You can go to the doctor and get your genome sequenced and find out how likely you are to get certain types of cancer. Maybe someday they'll also say, 'Hey, don't go to West Africa, it looks like you're susceptible to Ebola.' That's the dream."

"Or better yet," Rasmussen said, "Before you go to West Africa, you're going to need this vaccine."