Courtesy of Dietmar Schumucker
I got this cDNA sequence alignment for the Dscam gene in October 1999 while working in Larry Zipursky's lab. The gap in the alignment hinted at alternative splicing of the transmembrane domain. But that was just the tip of the iceberg: The handwritten inset is from my notes indicating not only this successful sequencing reaction but also that two reactions failed – much to my frustration (see the question marks). It turned out, however, that this was good news.
Serendipitously, I had designed sequencing primers that were located in the other variable regions of Dscam and therefore could only work in one cDNA. Based on this and the newly published Drosophila genome sequence, Jim Clemans, Huidy Shu, and myself were able to work out the mystery over the following few weeks. Six months later we reported1 that Dscam could be alternatively spliced into 38,016 different protein ...
