When billionaire philanthropist and entrepreneur Paul Allen, a cofounder of Microsoft, went looking for scientists to spend big on, he had one main rule: No sheep need apply.

“We wanted to find people who might not be where the herd is going,’’ said Tom Skalak, executive director of the new Paul G. Allen Frontiers Group, a $100 million initiative to back risky, cutting-edge science that more conventional funders might avoid.

The Frontiers Group, launched Wednesday, will start by bankrolling two Allen Discovery Centers; each will receive about $30 million over eight years.

The center at Stanford University, led by biologist Markus Covert, will combine computational modeling, bioengineering, cell biology, and immunology to understand how salmonella bacteria interact with the immune system and how drug resistance arises.

The center at Tufts University, led by biologist Michael Levin, aims to understand the “morphogenic code,’’ or the signals inside organisms that choreograph networks of cells into functional tissues and organs. Levin has found that electric fields and electric signaling between cells help give tissues, organs, and other body parts (like arms) their shapes. That’s called an “emergent’’ phenomenon because it cannot be predicted from genes and molecules or the other mainstays of reductionist biology.

“Our work is unique because we pay attention to higher [biological] levels than DNA and molecules,’’ Levin told STAT. “It’s a real challenge to get [grant] reviewers up to speed on that, and I’ve written way, way more proposals than have gotten funded.’’

The Allen money gives him a much needed cushion so he can do science rather than fund-raising, he said.

Allen’s initiative comes out of an eight-month listening tour in which Skalak and his team asked some 1,000 experts — senior scientists, junior scientists, policy wonks, and others — which promising ideas have been starved for support.

Some of the projects will probably flop. “But without risk, there is rarely significant reward, and unless we try truly novel approaches, we may never find the answers we seek,’’ Allen said in a statement ahead of the launch announcement at the National Academy of Sciences in Washington, D.C.

This isn’t the first time the billionaire has looked for unconventional scientists to back.

His Distinguished Investigators program, launched in 2010, funds scientists who take “out-of-the-box approaches at the very edges of knowledge,’’ as Allen put it. The program’s 2015 awards for Alzheimer’s research, for instance, largely passed over scientists who focus on the leading (and, so far, leading-to-nothing-useful-to-patients) hypothesis that the disease can be treated by eliminating amyloid plaques.

The new batch of Distinguished Investigators, receiving $1 million to $1.5 million each, includes scientists who have already rocketed to the tops of their fields.

Ethan Bier of the University of California, San Diego, will investigate where new body forms come from. Bier had been studying that mystery when he serendipitously hit on a way to rapidly spread edited genes through a population of fruit flies via a mechanism called gene drive.

Bier used the genome-editing system CRISPR-Cas9 to do that — and as it happens, CRISPR pioneer Jennifer Doudna of the University of California, Berkeley, was also named a Distinguished Investigator this year. CRISPR is part of bacteria’s antivirus defense, and Doudna will use her funding to look for other natural systems that can edit genomes. She’ll also be trying to change the behavior of cells by targeting RNA, in the hope of finding treatments for disease.

At MIT, biologist James Collins will use his $1.5 million to continue developing a way to see, in real time, the changes in bacteria genes that make them resistant to antibiotics.

Allen’s other major biomedical initiative is the eponymous Institute for Brain Science in Seattle, which he founded in 2003, right after his mother was diagnosed with Alzheimer’s. He has since given it $500 million. The Allen Institute has produced atlases of the mouse brain, composed of some 85 million images, and the human brain.

Sharon Begley can be reached at sharon.begley@statnews.com. Follow her on Twitter @sxbegle.