Blast ahead


nature genetics • volume 23 • december 1999 377 There we were, sixteen students and eight fullor part-time instructors in a notquite-finished empty building. Over the course of two weeks*, we were going to build four microarrayers from parts in boxes, amplify all 6,000 yeast genes using the polymerase chain reaction (PCR), use the microarrayers to spot the PCR products on glass slides, carry out four timecourse experiments, extract mRNA, hybridize the resulting labelled cDNAs to arrays, scan for fluorescence and analyse the results using a Perl database, installed by us on a Linux server. There was also a banquet. Building the arrayers was a blast. We bolted electrical and mechanical components onto a heavy aluminium platter, connected them together and to a motion controller card on a computer, and quickly obtained four more-or-less working robots. The “more-or-less” was the most interesting part of the course: on first pass, all of the robots had defects. Joe DeRisi, Pat Brown (Stanford University) and the other instructors provided hints on diagnosis, and we un-built, fixed and re-built until we got it right. Sometimes we’d just wired it wrongly, and felt stupid. In other cases, parts were partially or wholly defective. My favourite ‘blast-ahead’ moment was when a vacuum pump wouldn’t work. We took off the casing and found that the metal frame of the pump was bent, blocking the blades of the cooling fan. We hacked off the ends of the fan blades with the nearest sharp instrument, reassembled the pump and went on. Many defects were more subtle: a limit-detector magnet that rubbed against the motor casing and had to be ground down; two screws whose heads weren’t counter-sunk; a misaligned stage axis; a loose motor coupling. In most of these cases, we didn’t even know that there was a problem, but Joe could see or, more often, hear, that something wasn’t right, and would point it out. It was in this phase of testing, diagnosis, rebuilding and even re-machining that we came to understand the arrayers. By the third day, the arrayers were running beautifully and we moved on to the molecular biology part of the course. Much of this was familiar, although the logistics of running and analysing the products of 6,000 PCRs in a day and a half were interesting. The right microtitre dishes, multi-pipettors and gel boxes made this surprisingly easy. Some of the tricks for labelling cDNAs and making

DOI: 10.1038/70468

Cite this paper

@article{Futcher1999BlastA, title={Blast ahead}, author={Bruce Futcher}, journal={Nature Genetics}, year={1999}, volume={23}, pages={377-378} }