Uber for experiments

Print 08 December 2014
The Economist

A startup enables researchers to tap labs worldwide to conduct experiments on their behalf.

MOST research equipment is under-used. Once it has been budgeted for, grant proposals written or fee schedules set to cover its purchase, kit costing millions of dollars can sit idle for most of the working day. This inefficiency troubled Elizabeth Iorns, a biologist from New Zealand. So she came up with the idea of a marketplace where laboratories could rent out their machines to conduct experiments for others.

Dr Iorns started Science Exchange in 2011 when working as an assistant professor at the University of Miami. She was backed by Y Combinator, a Silicon Valley firm that helps startups, and she now serves as the exchange’s chief executive.

Dr Iorns is clear that certain laboratories are demonstrably better at some things than others. Her firm takes out contracts with some of the leading ones, including facilities at Johns Hopkins University, the Mayo Clinic and Harvard Medical School. It then provides ratings, reviews and other feedback, coupled with vetting, so that users can choose laboratories that can provide what they require and then compare pricing.

Unlike using a contract-research lab, which may involve lengthy negotiations and numerous visits to lawyers, Science Exchange gets each party to sign an agreement that governs all interactions and expectations for the work which will be carried out. This is especially valuable for startups as it saves time and money.

Ethan Perlstein, the founder of PerlsteinLab, has used both contract labs and Science Exchange. His San Francisco-based startup tries to find treatments for diseases that are too rare to attract the interest of big drugs firms. Many of these diseases are genetic. Although he intends to continue using contract labs for some of the company’s work, Dr Perlstein says the service negotiations involved “might add another month of lead time and another few thousand dollars of legal costs”.

In the first version of her exchange, Dr Iorns asked researchers to post their experimental needs and expected labs to reply with offers. This didn’t work. “Researchers are very private about the work they’re conducting,” she says. So the revised model gets labs to list their offerings. So far, this amounts to over 6,400 possible experiments. Researchers then obtain bids for the work they need or agree to posted fees. Science Exchange now has links to 1,000 labs and has handled $43.6m in quotations, $21.5m of that in the first nine months of 2014. The firm does not yet reveal the value of completed transactions, on which it levies a fee.

Each experiment may stand on its own. “It doesn’t require disclosure of the whole idea,” adds Dr Iorns, and participating labs make a legal commitment not to publish or to share any of the data they obtain. Having such work done by a third party can have advantages. It is harder for researchers to cherry-pick results unintentionally by shaping the process. There is also the benefit of having the work carried out by an experienced operator of the equipment concerned.

But some things researchers will want to keep in-house. As PerlsteinLab has grown it has begun buying a variety of equipment for “mission-critical work”. “There’s no way we’re going to outsource that,” says Dr Perlstein. Nor does he intend to rent his new gear out. But as some of it costs more than $500,000, one day he might be tempted.

Source

Return