Like most of his peers, Gunnar Carlsson spends his time thinking about hairy, theoretical math problems. It’s ivory tower stuff—he’s been a math professor for 30 years—which is just how the people in his field like it. “Mathematicians want to work on the deepest, hardest problems and get interesting intellectual results,” he says.
In 2008, Carlsson, while continuing his work at Stanford, co-founded Ayasdi, a Palo Alto tech startup. Ayasdi, which means “to seek” in Cherokee, is the first company to come out of Stanford’s math department and just received $10 million in funding from Khosla Ventures and Floodgate.
The company builds software that takes a complex branch of mathematics known as topology, the study of how shapes interact with space, and applies it to large volumes of data. People in fields as diverse as biotech, data security, and social networking believe the software could pull fresh insights out of huge databases in record time. “I view it as one of the real advances in data analysis to have arrived in the last 10 years,” says Eric Schadt, the director of the Institute for Genomics and Multiscale Biology at New York’s Mount Sinai Hospital, who has used the software to study bacterial outbreaks and genetic mutations.
With today’s powerful data analysis systems, users gather a ton of information—a breakdown of Wal-Mart Stores’ sales in the U.S. or things people “like” on Facebook—in one place and then run queries. The questioner typically comes in with a preconceived idea of what he’s looking for or at least a set of preconceived biases that determine the questions he asks.
The Ayasdi software, which customers including Merck and Raytheon have been testing for several months, runs dozens of algorithms and then illuminates patterns and relations between the data points. BN ImmunoTherapeutics, for example, has turned to the software for research help on Prostvac, a prostate cancer vaccine that is undergoing clinical trials. The researchers compare genetic markers, people’s ages, medical histories, and other factors to figure out which patients will most likely benefit from the vaccine. “In the past, we would form a hypothesis and say, ‘We think these three biomarkers are important,’ ” says Amanda Enstrom, a research scientist at BN ImmunoTherapeutics. “With Ayasdi, we really allow the data to show us what the important biomarkers are.”
The federal government has funded work in this area of mathematics for the last 10 to 15 years. At Stanford, Carlsson was part of a group of researchers that received money from Darpa, the research and development arm of the U.S. Department of Defense. The agency saw topology as promising for many applications, and no doubt helpful with national security investigations that require finding patterns among vast troves of information.
The startup’s software allows customers to upload their information from a website to Ayasdi’s data center, which applies the algorithms. The relationships between various data points get displayed as colorful, 3D pictures on the screen, and users can pose their queries via a Google-like search box. During one demonstration, Carlsson picks through genetic data on thousands of breast cancer patients and, with a couple of clicks, shows which groups of women will respond best to chemotherapy and what their DNA has in common.
Traditionally, drawing these types of correlations has taken years of painstaking work or been beyond the scope of today’s computing systems, says Mount Sinai’s Schadt. “It’s about taking hundreds of thousands of variables and scoring them across hundreds of thousands of people and trying to extract patterns,” he says. “We’re able to ask some novel questions.”
Ayasdi expects pharmaceutical, energy, and defense organizations will show the most interest in its technology. Enstrom, the research scientist, hopes to see the software used to analyze public health databases as scientists try to form a better understanding of the interplay between genes, environment, and lifestyle. “It may start better informing our growing field,” she says.