It’s fun to read association studies published in Cell; the molecular biology community generally takes a massively different path to an association than the current “big science” approach of massive genome-wide studies. Case in point: a recent paper identifying a non-synonymous variant in a previously unannotated gene associated with late-onset Alzheimer’s disease.

The approach the authors took was this: linkage studies had identified a region of chromosome 10 as potentially harboring a potential Alzheimer’s variant, and the hippocampus is among the first tissues affected by the disease. Thus, genes located in the chromosome 10 region and highly expressed in the hippocampus are potential candidates. There are a whole host of reasons you can come up with to convince yourself that this has no chance of working, but in this case, it did.

The authors identified a transcript that fit their profile, but of course, there was absolutely nothing known about it. So they painstakingly characterized it as a calcium channel and identified a common non-synonymous polymorphism in it that’s associated with Alzheimer’s in several different cohorts. To top it off, they show evidence that the SNP changes the activity of the channel. Overall, quite an impressive piece of work.