Arrays for the Masses—Setting up a Microarray Core Facility

Abstract

Early in the fall of 2001, the Oregon Health and Science University (OHSU) Spotted Microarray Core (SMC) was about to open for business when we noticed something odd about our cDNA arrays. We were printing arrays with 5700 genes, which took about 15 hours to complete. Probe DNA printed early in the print run stained brightly with the dye Syto61 and hybridized well to targets derived from human heart RNA. DNA printed after 7 hours was progressively fainter by both assays and any DNA printed after about 12 hours was completely absent from the slide. We ran numerous tests and came to the conclusion that the slides themselves were the problem. After a month of working with the slide manufacturer to solve the problem, which was supposedly unique, I posted a message to the microarray listserv about the problem. No one had any solutions, but I was inundated with e-mail from other cores having the same problem and not knowing what to do. Eventually, through a series of tests in our core, we learned two things: 1) most amino-silane slides had the same problem, but not all facilities experienced it, and 2) a new generation of slides coming onto the market no longer exhibited this problem.