This Sunday’s New York Times magazine has a rather whimsical and quixotic piece on synthetic biology. I was actually around at the beginning of this whole game as a postdoc in Jim Collin’s lab at Boston University in the mid-nineties. Back then Jim’s lab was mostly focused on posture control and stochastic resonance and I was transitioning into computational neuroscience with Nancy Kopell. One day in 1997 I think, Charles Cantor called a meeting with Jim, David Christini, who was a graduate student in Jim’s lab, and myself about using engineering methods to design bacteria. His example idea was to design a bacteria that could eat oil and then die. Jim, who was always open to new ideas, jumped right on it. I think the premise of the meeting was to present this proposal for a site visit by the Whitaker Foundation. This was when oil spills still dominated the news. I think we quickly cobbled something together and Jim presented the ideas. However, Dave was finishing up his thesis and heading off to Cornell in NYC and I was trying to learn about the hippocampus so the idea kind of sat idly for awhile.
Shortly afterwards, Jim recruited Tim Gardner, a first year grad student in biomedical engineering at BU, to work on the project. Jim got some lab space in Cantor’s lab and we started to talk about what we could do. At that time, I started reading about neural networks and noticed that the equations for a neural network looks remarkably like that for a gene regulatory network. My idea was to build a gene Hopfield network associative memory. I showed Tim how to use XPPAUT, a differential equation solver written by Bard Ermentrout, taught him a little about dynamical systems, and we started to write down equations. This was 1998.
I think we ended up with some circuit with tens if not twenty or more genes with complicated connections and so forth. Obviously, I was not cut out for this field because implementing this in bioware would have been a nightmare. In December of 1998, I spent a mini-sabbatical in Switzerland with Wulfram Gerstner. When I got back to the lab in January, Tim showed me his new idea, which was to make a genetic toggle switch. It was a circuit consiting of two mutually inhibiting genes, so that only one gene could be expressed at a time while suppressing the other and a stimulus, like tetracyclin, could knock the circuit from one gene expressed to the other. The idea was brilliant in its simplicity and other than showing Tim how to use XPPAUT and how not to muck things up with complications, I had no part of it. It may have been one of the biggest missed opportunities of my career. Tim then toiled away at this idea and it resulted in the celebrated Nature paper that kicked off the synthetic biology revolution. In that same issue, Mike Elowitz and Stan Leibler published the article on their genetic oscillator.
Erratrum: I think I am off by one year. I spent the mini-sabbatical in Switzerland in December 1997. So Tim also started in 1997 and the meeting with Cantor may have been 1996.