Skip to main content
. 2020 Apr 7;9:e53449. doi: 10.7554/eLife.53449

Figure 1. Association of burden of PTVs binned by population frequency (or minor allele frequency, MAF) with lifespan, healthspan, and parental age at death.

Number of ultra-rare variants belonging to each MAF bin was calculated for each exome and tested for association with lifespan phenotypes using Cox proportional hazards model and covariates to account for population structure. UKBBN lifespan was tested for associations with corresponding PTVs burdens using sex and 20 first principal components (PCs) from PCA analysis with 1000G project. UKB lifespan during follow-up was tested for associations using sex, age of enrollment, assessment centers and 40 PCs provided by UKB as covariates. UKB healthspan, mother’s and father’s ages at death were tested for associations using sex, assessment centers and 40 PCs as covariates. Beta coefficients estimated by Cox proportional hazards model (Cox PH beta) are plotted as dots with whiskers representing 95% confidence intervals. p-Values are shown for significant results only. Blue color designates statistically significant associations. Red dashed line designates zero Cox PH beta coefficient value. MAF - minor allele frequency, PTV - protein-truncating variants (defined as stop codon gains, frameshifts, canonical splice acceptor/donor sites variant), UKB - UK Biobank, UKBBN - UK Brain Bank Network.

Figure 1—source data 1. Source data for Figure 1.

Figure 1.

Figure 1—figure supplement 1. Distributions of PTV number per UKB exome depending on the variant population frequency (or minor allele frequency, MAF).

Figure 1—figure supplement 1.

The burden of variants increases with the frequency of the variant in the population. µ and σ shown in the upper right corners are mean and standard deviation of the corresponding distribution. PTV - protein-truncating variants (defined as stop codon gains, frameshifts, canonical splice acceptor/donor sites variant).