By Brad Stenger
Beyond Genomics has empty rooms and an empty building — a wise strategy for a startup
The managers of Beyond Genomics, a Waltham, Mass.-based startup, believe in having a plan. They’ve got a working-draft roadmap to take them from systems biology to value proposition. And if you’re a writer like me instead of a competitor, they’ll put the map on the wall and explain it to you. Kevin Barrett, executive VP, says, “This is our DNA. It drives what we do.”
Planning facilities and designing labs are two of the biggest bet-the-farm exercises any new organization has to go through. Working them into a systems approach has concrete benefits: less process guesswork, improved productivity, and reduced business risk. Chart a great map and it becomes a planning tool for facilities, biz strategy, informatics, and scientific discovery.
Of course, the problem with following your own map, no matter how elegantly conceived, is the same as with any document meant to navigate unknown territory: Does it really simulate the real world well enough to get you where you want to go?
Beyond Genomics’ systems biology roadmap guides the integration of disciplines. Well-defined roles are essential so that connections between groups are consistently meaningful. The roles also drive lab design — lab space zoned according to technical specialties furthers the purpose. “It allows a chromatographer to think like a chromatographer,” says Tom Londo, director of protein technologies. To tie together specialties, a lab designer or manager need only take his cue from the process map to obtain clear-cut guidance for work and material flows.
Room to grow in 4D
Headquarters is a fairly unassuming brick building, more blue-collar Boston than you might expect given the glitzy neighborhood nearby. A steady stream of truck traffic close outside the front door reinforces the low-key aura. The labs are places for smart, hardworking scientists, engineers, and computer jocks to be productive. That their space doubles as a showcase for leading-edge technology is incidental.
The main edifice has a twin — the company owns identical, side-by-side 40,000-square-foot buildings with 7,500-square-foot labs. Beyond Genomics occupies just 60 percent of one of those buildings. Architecturally, that means the swing spaces — areas that workgroups displace to when their current research space needs an update — have already been considered. With that buffer in place, the company can design on the fly to stay in sync with an ever-evolving system for biological research. The idea of continuous facilities improvement is a shift away from the rigid idea of master planning, but vital for a company that expects to make rapid progress.
In early October, the labs received a delivery from David Clemmer, a company co-founder and professor at Indiana University, of the first working prototype of a 4D mass spec. (The instrument gives two data dimensions based on electrospray, a reading of the time-of-flight ions inside the mass spec chamber, plus two new dimensions: a sophisticated reading of ion mobility and changes in relative position within the chamber). On the other side of a wall from mass spec #1 are empty sockets leading to an industrial-strength power supply, ready for Clemmer’s follow-on, which is scheduled for delivery in the second half of 2002.
At the moment, the data center fits into a typical closet — no need for elaborate design yet. Just off the wet labs and next to the mass specs, though, a surprising feature stands to make Beyond Genomics’ data center the envy of computational biologists. The previous owner of the building, a blood products company, needed enormous refrigerated rooms to maintain inventory. The chilly environment is conducive to overclocking computer chips. Manufacturers frown on the practice: as clockspeed goes up, overheating is a risk and chips can burn out. But a reduction in room temperature from, say, 70 degrees to 45 degrees could support a 50 percent bump-up in clockspeed. It seems that conditions are perfect for handling an explosion of data from the newfangled mass-spec instruments.
Brad Stenger is a freelance journalist who researches human-computer interaction in computational biology at the Georgia Institute of Technology, designs bioinformatic interfaces for Yale’s Gerstein Lab, and worked as a laboratory planner for architectural firm CUH2A. Send your comments to Brad at [email protected]