Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Letter to the Editor
  • Published:

MZ twin pairs or MZ singletons in population family-based GWAS? More power in pairs

Molecular Psychiatry volume 19, pages 1154–1155 (2014)Cite this article

Subjects

Family-based genome-wide association studies (GWAS) involve testing the genetic association of (many) genetic variants with the phenotype of interest, while taking into account the relatedness among family members. Occasionally in family-based GWAS, including monozygotic (MZ) twins, the data from one MZ twin are dropped, thus reducing the MZ pairs to singletons (for example, Loukola et al.,1 Lowe et al.,2 Parsons et al.3 and Psychosis Endophenotypes International Consortium et al.4). From a statistical power perspective, this practice is not optimal. To evaluate the issue of power, we consider the effective sample size (NE), that is, the number of independent cases that provides the same power as N MZ twin pairs. Given the MZ intraclass correlation of ρ, the effective sample size is calculated as NE=(2*N)/(1+ρ), where NE ranges from N (ρ=1) to 2*N (ρ=0). For instance, given N=1000 pairs, discarding data from one MZ twin reduces the sample size to 1000 singletons, that is, the NE assuming ρ=1. However, given ρ=0.2 (0.4, 0.7), the NE is 1667 (1429, 1176), so that 1000 twin pairs (2000 individuals) are equivalent—in terms of power—to 1667 (1429, 1176) unrelated individuals. To illustrate the loss in power, we consider a candidate gene explaining 1% of the variance, the power to detect the association in linear regression with N=1000 MZ twin pairs (α=0.001). MZ singletons, that is, 1000 unrelated subjects, provide a power of 0.450. Retaining data from both MZ twins (1000 pairs), the power varies with ρ as follows: 0.884 (ρ=0), 0.789 (ρ=0.2), 0.643 (ρ=0.5) and 0.519 (ρ=0.8). We refer to Supplementary Figure 1 for more details. Importantly, the gains associated with retaining MZ pairs involve no additional genotyping costs. That is, given the almost perfect concordance rate observed in MZ twins (>99%), genotyping one twin suffices in twins of confirmed monozygosity.

An important related question is whether retaining both MZ twins affects the type I error rate, that is, does the empirical type I error rate equal the chosen α? We checked the type I error rate by means of simulations. Our results indicate that the empirical type I error rate is correct, that is, invariably equals the chosen α (for details we refer to Supplementary Table 1 and to Supplementary Figure 2). Minică et al.5 evaluated the type I error in samples involving MZ twins, full sibs and parents, and also found that the empirical α closely resembled the nominal α. We conclude that the presence of MZ twins alone, or MZ twins in combination with other family members, does not affect the type I error rate.

This is a preview of subscription content, access via your institution

Relevant articles

Open Access articles citing this article.

Access options

Buy this article

  • Purchase on Springer Link
  • Instant access to full article PDF
Buy now

Prices may be subject to local taxes which are calculated during checkout

References

  1. Loukola A, Wedenoja J, Keskitalo-Vuokko K, Broms U, Korhonen T, Ripatti S et al. Mol Psychiatry 2014; 19: 615–624.

    Article  CAS  PubMed  Google Scholar 

  2. Lowe JK, Maller JB, Pe'er I, Neale BM, Salit J, Kenny EE et al. PLoS Genet 2009; 5: e1000365.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Parsons MJ, Lester KJ, Barclay NL, Nolan PM, Eley TC, Gregory AM . Am J Med Genet Part B: Neuropsychiatr Genet 2013; 162: 431–438.

    Article  CAS  Google Scholar 

  4. Psychosis Endophenotypes International Consortium, Wellcome Trust Case-Control Consortium, Bramon E, Pirinen M, Strange A, Lin K et al. Biol Psychiatry 2014; 75: 386–397.

    Article  PubMed Central  Google Scholar 

  5. Minica CC, Dolan CV, Kampert MM, Boomsma DI, Vink JM . Eur J Hum Genet 2014; doi:10.1038/ejhg.2014.94 (e-pub ahead of print).

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Rietveld CA, Medland SE, Derringer J, Yang J, Esko T, Martin NW et al. Science 2013; 340: 1467–1471.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  7. Anttila V, Winsvold BS, Gormley P, Kurth T, Bettella F, McMahon G et al. Nat Genet 2013; 45: 912–917.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Codd V, Nelson CP, Albrecht E, Mangino M, Deelen J, Buxton JL et al. Nat Genet 2013; 45: 422–427.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  9. Ripke S, Wray NR, Lewis CM, Hamilton SP, Weissman MM, Breen G et al. Mol Psychiatry 2013; 18: 497–511.

    Article  CAS  PubMed  Google Scholar 

  10. Carey VJ, Lumley T, Ripley B . 2012. http://CRAN.R-project.org/package=gee.

Download references

Acknowledgements

Camelia C Minică and Jacqueline M Vink are supported by the ERC starting grant 284167. Conor V Dolan is supported by the European Research Council (Genetics of Mental Illness; grant number: ERC-230374). The statistical analyses were carried out on the Genetic Cluster Computer (http://www.geneticcluster.org), which is supported by the Netherlands Scientific Organization (NWO 480-05-003), the Dutch Brain Foundation and the Department of Psychology and Education of the VU University Amsterdam.

Author information

Authors and Affiliations

  1. Department of Biological Psychology, VU University Amsterdam, Amsterdam, The Netherlands

    C C Minică, D I Boomsma, J M Vink & C V Dolan

Authors
  1. C C Minică
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. D I Boomsma
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. J M Vink
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. C V Dolan
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to C C Minică.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Molecular Psychiatry website

Supplementary information

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Minică, C., Boomsma, D., Vink, J. et al. MZ twin pairs or MZ singletons in population family-based GWAS? More power in pairs. Mol Psychiatry 19, 1154–1155 (2014). https://doi.org/10.1038/mp.2014.121

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1038/mp.2014.121

This article is cited by

Search

Advanced search

Quick links