Fig. 1: Genome-wide association study on common variants.

Association studies were performed using logistic regression with allele dosage as the predictor assuming log-additive allele effects adjusted for relevant covariates. Raw two-sided P are reported as −log10(P). A Manhattan plot of the FTLD-TDP All patients versus controls association study (inflation, λ = 1.05). B Manhattan plot of the FTLD-TDP A patients versus controls association study (λ = 1.07). C Manhattan plot of the FTLD-TDP B patients versus controls association study (λ = 1.06). D Manhattan plot of the FTLD-TDP C patients versus controls association study (λ = 1.05). The red-dotted line represents the genome-wide significance level (p = 5 × 10−8).