Jesse Bloom is fascinated by evolution.

Think how amazing it is, he says, that in nature, small, random genetic changes can add up to something new and wonderful – a stark contrast to the world of things. “If you make random mutations to your car with a baseball bat, you’re probably not going to improve the car very much,” he says. “So what is it about biology that allows living things to be so good at evolving?”

Bloom is tackling this question by helping to pioneer new approaches for studying evolution. He’s combining biology with computational wizardry. “Traditionally, evolutionary biology has been an observational science,” he explains. Researchers looked back on the creation of new species, like Darwin’s finches, or bold new innovations, like the ability to fly, then searched for the genetic steps that paved the way.

But thanks in part to Bloom, scientists don’t have to wait for nature to do its work. “Now we can do large-scale experiments in the lab that can help to explain evolution,” Bloom says. In his own lab, he’s making tens of thousands of mutations in flu viruses to see what effect they have. At the same time, he’s also charting genetic changes that appear in nature. Bloom’s goal is to discern patterns in evolution and use his mathematical modeling skills to predict what’s coming – offering a major public health advance.

His work is paying off. In one recent discovery, Bloom and his collaborators uncovered how the flu virus managed to develop resistance to the drug Tamiflu. One key mutation prevents the drug from attacking flu viruses. But that genetic change, by itself, is actually lethal to the virus because it causes the virus to break apart, Bloom discovered. Several other genetic alterations shore up the virus, letting it survive the drug-blocking mutation.

Another mystery has been the sequence of genetic steps that flu viruses take as they evolve from year to year. Bloom tackled this question by analyzing genetic changes in viruses in immunocompromised patients, whose flu infections often last for months. Not only do these viruses mutate rapidly, he found, but the changes surprisingly parallel what happens to viruses out in the world.

The similarity suggests that, with clever mathematical sleuthing, scientists can get better at guessing what new strains humanity will face in coming years. “I don’t think we will ever be able to completely forecast the future, but it will be interesting to see how good we can get,” says Bloom.

He is uniquely suited to the task. Growing up in Hamilton, Montana, Bloom loved all types of science, including biology, physics, chemistry, math, and computer programming. The problem was picking just one. Fortunately, at the time, biologists were beginning to explore the power of computers, so Bloom was able to combine his quantitative bent with his fascination for evolution.

After studying protein folding as a University of Chicago undergraduate, Bloom wanted to do something more mathematical, such as using his modeling expertise to predict proteins’ shapes from their sequence. That turned out to be fiendishly difficult, “so I thought maybe I can study how they evolved,” he says. That led him to the California Institute of Technology as a PhD student and postdoc, working to harness evolution to engineer new and useful enzymes. There, “I learned I was more of a curiosity-driven biologist than a problem-solving engineer,” he says.

Now, that curiosity – backed up by rigorous computation – may finally bring answers to some of the great questions of evolution.