It was my privilege this past weekend to moderate a lively Medicine X panel comprising three speakers who'd just given presentions on different phases of the avalanche of information already rumbling down the mountain of scientific curiosity. Sean Bonner talked about his nonprofit company Safecast's efforts to put into place myriad mobile sensors to monitor radiation in the wake of the Japanese reactor disaster. Stanford medical data king and father of several biomedical start-ups Atul Butte, MD, PhD, described how his team mined public databases to isolate a protein that's likely a key player (and until now an entirely overlooked one) in the development of Type 2 diabetes. Pete Binfield, PhD, former publisher of the open-access journal PLoS ONE, explained how his latest venture, PeerJ will push the open-access envelope even further.
I had a bone to pick concerning the quality of the information that’s being scooped up, sifted, sorted, and spewed out in ever-increasing quantities. A few weeks ago I got a notice in the mail informing me a traffic violation on my part had been picked up by an automatic tracking device. Seems I'd evaded a two-dollar toll on Highway 261 on Sept. 10, at 6:50 a.m.
But Sept. 10 was a Monday. On Mondays I work from my home in San Francisco and don't open my eyes till 6:30. Highway 261 is in Orange County, easily 500 miles away. So, say I flew out of bed at 6:30 sharp, skipped breakfast and brushing my teeth, raced down the stairs, and jumped in my car ... I’d still have to drive really fast (like, 2,000 miles an hour) to get to that toll booth within 20 minutes. Assuming I obeyed even half of the traffic signals on my way out of town, we’re talking relativistic speeds.
And I got popped for jumping a crummy two-dollar toll, but not for speeding?
I knew something was fishy. So I called the 800 number and they checked their photo and, sure enough, they were off by one digit on the license-plate number. They were very nice about it, too – which tells me this happens a lot. It’s an example of what people in the bioscience biz call a “false positive.”
And it got me thinking about large-scale data collection and crunching. So I asked Sean, Atul, and Pete how their respective approaches would safeguard or improve quality. Their unanimous prescription, captured in this video (our session starts at about 02:59:50), in a word: transparency.
Previously: Mining medical discoveries from a mountain of ones and zeroes, The data deluge: A report from Stanford Medicine magazine and Stanford's Atul Butte discusses outsourcing research online at TEDMED
Photo by plushev
More news about Stanford Medicine X is available in the Medicine X category.
