First GWAS: SNCA, MAPT, then 24 loci

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Javier Simón-Sánchez, Andrew Singleton, John Hardy and Thomas Gasser (last author) and colleagues (European scan, with NIH intramural, Medical Research Council and Parkinson's UK grants); Wataru Satake, Tatsushi Toda (last author) and colleagues (Japanese scan); Mike Nalls and colleagues of the International Parkinson's Disease Genomics Consortium, Andrew Singleton last author (2014; grants from NIH, Wellcome Trust, Medical Research Council and Parkinson's UK, among others). All in Nature Genetics. Funders are from Europe PMC grant records.

What happened

A genome-wide association study (GWAS) compares millions of common DNA variants in people with a disease and in healthy controls. It needs no prior guess about which gene matters.

• 2009, Europe. The Simón-Sánchez scan (1,713 cases, 3,978 controls, then 3,361 cases and 4,573 controls for replication) found SNCA (odds ratio 1.23) and MAPT (0.77), each at P near 10^-16. It is described in landmark 125. Nat Genet 41:1308–1312.
• 2009, Japan. Satake et al. ran a scan and two replications in 2,011 cases and 18,381 controls. They found a new locus on chromosome 1q32, named PARK16 (P = 1.5 x 10^-12), and a second new one at BST1 (P = 3.9 x 10^-9). SNCA (P = 7.4 x 10^-17) and LRRK2 (P = 2.7 x 10^-8) were also strongly linked. Compared with the European scan, PARK16, SNCA and LRRK2 were shared across populations. BST1 and MAPT differed between them.
• 2014, pooled. Nalls et al. combined earlier scans: 13,708 cases and 95,282 controls, over 7.9 million variants. Twenty-six loci reached genome-wide significance. With a replication set of 5,353 cases and 5,551 controls, 24 loci held up, including six new ones (SIPA1L2, INPP5F, MIR4697, GCH1, VPS13C, DDRGK1). Counting second signals within four loci, that made 28 independent risk variants. Each variant's effect is small. People in the top fifth of combined genetic risk had 3.3 times the odds of disease of those in the bottom fifth. Nat Genet 46:989–993.

Why it moved us toward a cure

The scans showed that the genes behind rare inherited forms (SNCA, landmark 4; LRRK2) also carry common risk variants in ordinary cases. That makes those proteins plausible targets for most patients, not only for rare families. The 2014 list added signals from dopamine synthesis and lysosome biology, and it set up the larger 2019 map of 90 signals (landmark 19).

Outcome

Advanced. Later, larger analyses superseded the counts. The 2009 scans were the proof that the method works for Parkinson's, and the two 2009 scans, run by separate teams in Europe and Japan, agreed on SNCA, PARK16 and LRRK2 (MAPT and BST1 differed by population).

Sources

• Satake et al., Nat Genet 2009;41:1303–1307, DOI 10.1038/ng.485
• Nalls et al., Nat Genet 2014;46:989–993, DOI 10.1038/ng.3043 (open access, PMC4146673)
• Simón-Sánchez et al., Nat Genet 2009, DOI 10.1038/ng.487
• Not yet verified: the PMC table header gives 13,728 cases against 13,708 in the abstract and text (the abstract figure is used). Europe PMC lists no erratum or retraction for the three papers; the two 2009 papers share one commentary (Nat Genet 41:1261–1262).