Maria Ban, An Goris, Åslaug R Lorentzen, Amie Baker, Tania Mihalova, Gillian Ingram, David R Booth, Robert N Heard, Graeme J Stewart, Elke Bogaert, Bénédicte Dubois, Hanne F Harbo, Elisabeth G Celius, Anne Spurkland, Richard Strange, Clive Hawkins, Neil P Robertson, Frank Dudbridge, James Wason, Philip L De Jager, David Hafler, John D Rioux, Adrian J Ivinson, Jacob L McCauley, Margaret Pericak-Vance, Jorge R Oksenberg, Stephen L Hauser, David Sexton, Jonathan Haines, Stephen Sawcer

In a recent genome-wide association study (GWAS) based on 12,374 non-synonymous single nucleotide polymorphisms we identified a number of candidate multiple sclerosis susceptibility genes. Here, we describe the extended analysis of 17 of these loci undertaken using an additional 4234 patients, 2983 controls and 2053 trio families. In the final analysis combining all available data, we found that evidence for association was substantially increased for one of the 17 loci, rs34536443 from the tyrosine kinase 2 (TYK2) gene (P=2.7 x 10(-6), odds ratio=1.32 (1.17-1.47)). This single nucleotide polymorphism results in an amino acid substitution (proline to alanine) in the kinase domain of TYK2, which is predicted to influence the levels of phosphorylation and therefore activity of the protein and so is likely to have a functional role in multiple sclerosis.