Cambridge startup’s software to help in battle against cancer

In In The News by Barbara Jacoby

[pro_ad_display_adzone id=”443489″ ajax_load=”1″ flyin=”1″ flyin_position=”bottom-center” flyin_delay=”1″] By: Jordan Graham


Trying to cure cancer means a lot of time, money and researchers, but Vice President Joe Biden’s cancer “moonshot” initiative also will mean mountains of data for researchers to pore over.

Tamr, a Cambridge startup that makes big data software, said last week that it would give its software to any researchers who are part of Biden’s effort.

“This is not about making money for us, this is about saving lives,” said Nidhi Aggarwal, head of product and strategy at Tamr. “We do believe this does require a moonshot-like effort, it is something that does require collaboration like we’ve never seen before.”

Biden announced the initiative last year with a promise to increase federal funding for cancer research by 20 percent, as well as foster collaboration between federal agencies, industry, researchers and patient groups. In announcing the effort, Biden emphasized the need for data to be available to more researchers.

“The science, data, and research results are trapped in silos, preventing faster progress and greater reach to patients. It’s not just about developing game-changing treatments — it’s about delivering them to those who need them,” Biden said in a blog post earlier this year.

That gap is where Tamr thinks it can help make the data more useful to researchers.

Tamr’s software helps companies bring data from multiple sources into the same place and in the same format to make it easier to analyze. The software can, for example, understand that two different spellings of a genome are the same and automatically correct them, and learn about similar discrepancies without being told.

“One thing we’ve seen with these previous projects is there is a huge task involved in the integration and aggregation of this data,” said Timothy Danford, a field engineer at Tamr.

Tamr is not meant exclusively for health care companies, but many at the company — from the CEO on down — have experience in the health care world. Danford spent time at the Broad Institute developing software for cancer researchers. Its customers include Novartis, but also GE and Thompson Reuters.

Aggarwal said scientists at companies spent as much as 90 percent of the time manually inputing and converting data.

“In some companies this is called the dung beetle patrol,” she said.

John Halamka, chief information officer at Beth Israel Deaconess Medical Center, said big data could mean big improvements in health care.

When his wife was diagnosed with cancer several years ago, Halamka turned to big data tools he helped develop at Harvard. That information, based on results from women with similar characteristics, called for a specific kind of chemotherapy instead of the standard surgery.

“The hope is that the detailed experience of others plus the lifetime record of the patient being treated can be combined to deliver precision medicine, improving quality while reducing cost,” he said. “Big data enables clinicians to understand the experience of past similar patients, including treatment plans and outcomes.”