Subscribe to Journal
Get full journal access for 1 year
only $4.92 per issue
All prices are NET prices.
VAT will be added later in the checkout.
Tax calculation will be finalised during checkout.
Rent or Buy article
Get time limited or full article access on ReadCube.
All prices are NET prices.
All the data used in this study are available with the permission of the UK Biobank.
Samson, M. et al. Resistance to HIV-1 infection in Caucasian individuals bearing mutant alleles of the CCR-5 chemokine receptor gene. Nature 382, 722–725 (1996).
Hütter, G. et al. Long-term control of HIV by CCR5 delta32/delta32 stem-cell transplantation. N. Engl. J. Med. 360, 692–698 (2009).
Gupta, R. K. et al. HIV-1 remission following CCR5Δ32/Δ32 haematopoietic stem-cell transplantation. Nature 568, 244–248 (2019).
Wei, X. & Nielsen, R. Retraction note: CCR5-∆32 is deleterious in the homozygous state in humans. Nat. Med. 25, 1796 (2019).
Wei, X. & Nielsen, R. Deviations from Hardy Weinberg equilibrium at CCR5-Δ32 in large sequencing data sets. Preprint at bioRxiv https://www.biorxiv.org/content/10.1101/768390v2 (2019).
Harrison, S. “CCR5-∆32 is deleterious in the homozygous state in humans”—is it? Sean Harrison: Blog https://seanharrisonblog.com/2019/06/20/ccr5-%e2%88%8632-is-deleterious-in-the-homozygous-state-in-humans-is-it/ (2019).
Gudbjartsson, D. et al. CCR5-del32 is not deleterious in the homozygous state in humans. Preprint at bioRxiv https://doi.org/10.1101/788117 (2019).
Karczewski, K. J., Gauthier, L. D. & Daly, M. J. Technical artifact drives apparent deviation from Hardy–Weinberg equilibrium at CCR5-∆32 and other variants in gnomAD. Preprint at bioRxiv https://doi.org/10.1101/784157 (2019).
Tanigawa, Y. & Rivas, M. A. Reported CCR5-∆32 deviation from Hardy-Weinberg equilibrium is explained by poor genotyping of rs62625034. Preprint at bioRxiv https://doi.org/10.1101/791517 (2019).
This research has been conducted using the UK Biobank Resource under application no. 31063. We acknowledge the participants in the UK Biobank. We are grateful to B. Neale and A. Price for critical comments, and to S. Harrison for a blog posting that showed how the association results and HWE P values at rs113010081 were qualitatively discordant with those at rs62625034, which prompted us to re-examine these issues. We thank K. Karczewski, K. Stefansson and M. Daly for sharing with us early versions of two other manuscripts re-examining the evidence of association to mortality at CCR5, and working with us to post all manuscripts together. This work was funded in part by NIH grants GM100233 and HG006399, the Paul Allen Family Foundation, the John Templeton Foundation (grant 61220) and the Howard Hughes Medical Institute.
The authors declare no competing interests.
Peer review information Kate Gao was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Survival rates for individuals with 0, 1, or 2 copies of the rare allele or No Call (NC) for variants tagging the CCR5-∆32 deletion. First row: Cumulative survival rates. Numbers are one-sided p-values of a Cox proportional hazard model which compares survival rates of individuals with 0 or 1 alleles to those with 2 alleles. Second row: non-cumulative survival rates. Third row: Number of individuals who have died in any given year with 2 copies of rare allele (see also Supplementary Table 7).
Confusion matrix for different markers with missing data. The last column of the first panel shows that individuals with missing genotype at rs62625034 are enriched for ∆32/∆32 according to rs333_sequenced. This can lead to a violation of HWE at rs62625034. All white British samples of UK Biobank WES data shared with UK Biobank Axiom array data are used in this figure.
Simulated HWE Chi-squared p-values at two variants with minor allele frequency of 11% with r2 of 0.95, in a sample of 400,000 individuals. Both variants are initially in HWE. We then remove a subset of samples which are homozygous for the rare allele at SNP 1. This leads to a deviation from HWE at SNP 1, but it also leads to a similar deviation from HWE at SNP 2. Only simultaneous selection acting in the opposing direction on SNP 2, or technical artifacts which create a dependence of missingness in one SNP on genotype in the other SNP explain a situation where HWE p-values are very different at both SNPs. Error bars denote the 5th and 95th percentile out of 100 replicates in each bin.
Power to detect effects on mortality of a genotype with the frequency of ∆32/∆32 in a sample of the same total size and mortality rate as the cohort studied here, as a function of relative risk. The power to detect a 20% increase in mortality rate at a 0.05 significance level is 75%.
Odds ratios (e𝛽) for all case-control phenotypes in five variants as a function of sample prevalence. Colors represent uncorrected p-values. Open circles represent case-control phenotypes with 10 or fewer cases in ∆32/∆32 individuals. Only phenotypes with more than five cases in ∆32/∆32 individuals are shown.
QQ-plot of the associations across all phenotypes. Each variant is plotted in a different color. Only phenotypes with more than five cases in ∆32/∆32 individuals are shown.
About this article
Cite this article
Maier, R., Akbari, A., Wei, X. et al. No statistical evidence for an effect of CCR5-∆32 on lifespan in the UK Biobank cohort. Nat Med 26, 178–180 (2020). https://doi.org/10.1038/s41591-019-0710-1