Ultra-rare disruptive and damaging mutations influence educational attainment in the general population

Journal name:
Nature Neuroscience
Volume:
19,
Pages:
1563–1565
Year published:
DOI:
doi:10.1038/nn.4404
Received
Accepted
Published online

Disruptive, damaging ultra-rare variants in highly constrained genes are enriched in individuals with neurodevelopmental disorders. In the general population, this class of variants was associated with a decrease in years of education (YOE). This effect was stronger among highly brain-expressed genes and explained more YOE variance than pathogenic copy number variation but less than common variants. Disruptive, damaging ultra-rare variants in highly constrained genes influence the determinants of YOE in the general population.

At a glance

Figures

  1. Association between number of disruptive, damaging and synonymous URVs in HC genes and YOE.
    Figure 1: Association between number of disruptive, damaging and synonymous URVs in HC genes and YOE.

    Disruptive and damaging URVs but not synonymous URVs are significantly associated with reduced YOE. The size of the squares is proportional to the size of the study. Bars are 95% confidence intervals. All estimates were obtained from a linear regression model. Meta-analysis results were obtained using a fixed-effect approach.

  2. Association between numbers of disruptive, damaging and synonymous URVs for different gene sets.
    Figure 2: Association between numbers of disruptive, damaging and synonymous URVs for different gene sets.

    The intersection between HC and brain-expressed genes yields the strongest reduction in YOE. We only report the meta-analysis results (n = 14,133). Bars are 95% confidence intervals. All estimates were obtained from a linear regression model and combined using fixed-effect meta-analysis.

  3. Association between each of the normalized scores (polygenic, runs of homozygosity, URVs and pathogenic CNVs) and YOE.
    Figure 3: Association between each of the normalized scores (polygenic, runs of homozygosity, URVs and pathogenic CNVs) and YOE.

    The results presented are from meta-analysis of Swedish WES, Estonian WGS and Finnish WGS studies (n = 13,353), except for the polygenic score, which is calculated only in the Swedish WES study (n = 10,651). Note that we plotted 1 polygenic score to obtain a negative association with YOE. The horizontal bars represent 95% confidence intervals. All the estimates were obtained from a linear regression model and combined using fixed-effect meta-analysis.

  4. First three principal components for each study
    Supplementary Fig. 1: First three principal components for each study
  5. Association between disruptive, damaging and synonymous URVs and YOE in cases and controls of schizophrenia.
    Supplementary Fig. 2: Association between disruptive, damaging and synonymous URVs and YOE in cases and controls of schizophrenia.

    Similar associations are observed in cases and control of schizophrenia. Total individuals included N=14,133.

  6. Association between disruptive, damaging and synonymous URVs and YOE after excluding individuals with neurodevelopmental disorders.
    Supplementary Fig. 3: Association between disruptive, damaging and synonymous URVs and YOE after excluding individuals with neurodevelopmental disorders.

    Only Swedish non-schizophrenic participants are included in this analysis. Similar associations are observed once individuals with neurodevelopmental disorders are excluded.

  7. Association between disruptive and damaging URVs with YOE for different quantiles of brain gene-expression.
    Supplementary Fig. 4: Association between disruptive and damaging URVs with YOE for different quantiles of brain gene-expression.

    Only Swedish participants are included (N=10,651).

  8. Association between disruptive, damaging and synonymous URVs and YOE in highly brain-specific HC genes, sparsely brain-specific HC genes and all highly brain-specific genes.
    Supplementary Fig. 5: Association between disruptive, damaging and synonymous URVs and YOE in highly brain-specific HC genes, sparsely brain-specific HC genes and all highly brain-specific genes.

    Meta-analysis results (N=14,133).

  9. Association between polygenic score and YOE, in individuals carrying 0 or [ge]1 URVs or pathogenic CNVs
    Supplementary Fig. 6: Association between polygenic score and YOE, in individuals carrying 0 or ≥1 URVs or pathogenic CNVs

    Only Swedish participants are included (N=10,651).

  10. Q-Q plots for the gene-based burden test for association between disruptive and damaging variants and YOE.
    Supplementary Fig. 7: Q-Q plots for the gene-based burden test for association between disruptive and damaging variants and YOE.

    Each point is a gene. Results are from meta-analysis across studies (N=14,133). In the upper panels we included only URVs, in the lower panels we included variants with minor allele frequency < 0.05%.

References

  1. McLendon, M.K. & Perna, L.W. Ann. Am. Acad. Pol. Soc. Sci. 655, 615 (2014).
  2. Haveman, R. & Wolfe, B. J. Econ. Lit. 33, 18291878 (1995).
  3. Krapohl, E. et al. Proc. Natl. Acad. Sci. USA 111, 1527315278 (2014).
  4. Cutler, D.M. & Lleras-Muney, A. Education and health: evaluating theories and evidence. National Bureau of Economic Research Working Paper Series 12352 (2006).
  5. Okbay, A. et al. Nature 533, 539542 (2016).
  6. Marioni, R.E. et al. Intelligence 44, 2632 (2014).
  7. Branigan, A.R., McCallum, K.J. & Freese, J. Soc. Forces 92, 109140 (2013).
  8. Zuk, O. et al. Proc. Natl. Acad. Sci. USA 111, E455E464 (2014).
  9. Lek, M. et al. Nature 536, 285291 (2016).
  10. Genovese, G. et al. Nat. Neurosci. http://dx.doi.org/10.1038/nn.4402 (2016).
  11. Gilissen, C. et al. Nature 511, 344347 (2014).
  12. Iossifov, I. et al. Nature 515, 216221 (2014).
  13. Robinson, E.B. et al. Nat. Genet. 48, 552555 (2016).
  14. GTEx Consortium. Nat. Genet. 45, 580585 (2013).
  15. Joshi, P.K. et al. Nature 523, 459462 (2015).
  16. Stefansson, H. et al. Nature 505, 361366 (2014).
  17. Wu, M.C. et al. Am. J. Hum. Genet. 89, 8293 (2011).
  18. Davies, K. Br. J. Nurs. 15, 252256 (2006).
  19. Yang, J. et al. Nat. Genet. 47, 11141120 (2015).
  20. Ripke, S. et al. Nat. Genet. 45, 11501159 (2013).
  21. Leitsalu, L. et al. Int. J. Epidemiol. 44, 11371147 (2015).
  22. Vartiainen, E. et al. Int. J. Epidemiol. 39, 504518 (2010).
  23. Jostins, L. et al. Nature 491, 119124 (2012).
  24. Schizophrenia Working Group of the Psychiatric Genomics Consortium. Nature 511, 421427 (2014).
  25. DePristo, M.A. et al. Nat. Genet. 43, 491498 (2011).
  26. Li, H. Bioinformatics 30, 28432851 (2014).
  27. Cingolani, P. et al. Front. Genet. 3, 35 (2012).
  28. Cingolani, P. et al. Fly (Austin) 6, 8092 (2012).
  29. Liu, X., Jian, X. & Boerwinkle, E. Hum. Mutat. 34, E2393E2402 (2013).
  30. Samocha, K.E. et al. Nat. Genet. 46, 944950 (2014).
  31. Chang, C.C. et al. Gigascience 4, 7 (2015).
  32. Marshall, C. et al. Preprint at bioRxiv http://dx.doi.org/10.1101/040493 (2016).
  33. Handsaker, R.E. et al. Nat. Genet. 47, 296303 (2015).
  34. Madsen, B.E. & Browning, S.R. PLoS Genet. 5, e1000384 (2009).
  35. Ionita-Laza, I., Lee, S., Makarov, V., Buxbaum, J.D. & Lin, X. Am. J. Hum. Genet. 92, 841853 (2013).

Download references

Author information

  1. These authors contributed equally to this work.

    • Andrea Ganna &
    • Giulio Genovese

Affiliations

  1. Analytic and Translational Genetics Unit, Massachusetts General Hospital, Boston, Massachusetts, USA.

    • Andrea Ganna,
    • Daniel P Howrigan,
    • Andrea Byrnes,
    • Mitja I Kurki,
    • Alex Bloemendal,
    • Jonathan M Bloom,
    • Jacqueline I Goldstein,
    • Timothy Poterba,
    • Cotton Seed,
    • Elise B Robinson,
    • Mark J Daly,
    • Aarno Palotie &
    • Benjamin M Neale
  2. Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.

    • Andrea Ganna,
    • Giulio Genovese,
    • Daniel P Howrigan,
    • Andrea Byrnes,
    • Mitja I Kurki,
    • Seyedeh M Zekavat,
    • Christopher W Whelan,
    • Alex Bloemendal,
    • Jonathan M Bloom,
    • Jacqueline I Goldstein,
    • Timothy Poterba,
    • Cotton Seed,
    • Robert E Handsaker,
    • Pradeep Natarajan,
    • Elise B Robinson,
    • Sekar Kathiresan,
    • Mark J Daly,
    • Steven A McCarroll,
    • Tõnu Esko &
    • Benjamin M Neale
  3. Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, Massachusetts, USA.

    • Andrea Ganna,
    • Giulio Genovese,
    • Daniel P Howrigan,
    • Andrea Byrnes,
    • Mitja I Kurki,
    • Christopher W Whelan,
    • Alex Bloemendal,
    • Jonathan M Bloom,
    • Jacqueline I Goldstein,
    • Timothy Poterba,
    • Cotton Seed,
    • Robert E Handsaker,
    • Diane Gage,
    • Elise B Robinson,
    • Mark J Daly,
    • Steven A McCarroll,
    • Aarno Palotie &
    • Benjamin M Neale
  4. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden.

    • Andrea Ganna,
    • Patrick F Sullivan &
    • Christina M Hultman
  5. Department of Genetics, Harvard Medical School, Boston, Massachusetts, USA.

    • Giulio Genovese,
    • Christopher W Whelan,
    • Robert E Handsaker &
    • Steven A McCarroll
  6. Institute for Molecular Medicine Finland, FIMM, University of Helsinki, Helsinki, Finland.

    • Mitja I Kurki &
    • Aarno Palotie
  7. Center for Human Genetic Research and Cardiovascular Research Center, Massachusetts General Hospital, Boston, Massachusetts, USA.

    • Seyedeh M Zekavat,
    • Pradeep Natarajan &
    • Sekar Kathiresan
  8. Estonian Genome Center, University of Tartu, Tartu, Estonia.

    • Mart Kals,
    • Reedik Mägi,
    • Andres Metspalu &
    • Tõnu Esko
  9. Institute of Mathematics and Statistics, University of Tartu, Tartu, Estonia.

    • Mart Kals
  10. Department of Biological Psychology, VU University Amsterdam, Amsterdam, the Netherlands.

    • Michel G Nivard
  11. Department of Health, THL-National Institute for Health and Welfare, Helsinki, Finland.

    • Veikko Salomaa &
    • Jaana Suvisaari
  12. Division of Psychiatric Genomics, Department of Psychiatry, Icahn School of Medicine at Mount Sinai, New York, New York, USA.

    • Shaun M Purcell &
    • Pamela Sklar
  13. Institute for Genomics and Multiscale Biology, Icahn School of Medicine at Mount Sinai, New York, New York, USA.

    • Shaun M Purcell &
    • Pamela Sklar
  14. Departments of Genetics and Psychiatry, University of North Carolina, Chapel Hill, North Carolina, USA.

    • Patrick F Sullivan

Contributions

A.G. and G.G. designed the study, performed the analysis and wrote the manuscript. B.M.N. supervised the project. D.H., A. Byrnes, M.I.K., S.M.Z., C.W.W., M.K., M.G.N., P.N. and R.M. performed the analyses. A. Bloemendal, J.M.B., J.I.G., T.P., C.S. and R.E.H. developed and provided computational tools. D.G. provide data management support. E.B.R., A.M., V.S., J.S., S.M.P., P.S., S.K., M.J.D., S.A.M., P.F.S., A.P., T.E. and C.M.H. collected and provided the data.

Competing financial interests

The authors declare no competing financial interests.

Corresponding author

Correspondence to:

Author details

Supplementary information

Supplementary Figures

  1. Supplementary Figure 1: First three principal components for each study (116 KB)
  2. Supplementary Figure 2: Association between disruptive, damaging and synonymous URVs and YOE in cases and controls of schizophrenia. (77 KB)

    Similar associations are observed in cases and control of schizophrenia. Total individuals included N=14,133.

  3. Supplementary Figure 3: Association between disruptive, damaging and synonymous URVs and YOE after excluding individuals with neurodevelopmental disorders. (50 KB)

    Only Swedish non-schizophrenic participants are included in this analysis. Similar associations are observed once individuals with neurodevelopmental disorders are excluded.

  4. Supplementary Figure 4: Association between disruptive and damaging URVs with YOE for different quantiles of brain gene-expression. (79 KB)

    Only Swedish participants are included (N=10,651).

  5. Supplementary Figure 5: Association between disruptive, damaging and synonymous URVs and YOE in highly brain-specific HC genes, sparsely brain-specific HC genes and all highly brain-specific genes. (50 KB)

    Meta-analysis results (N=14,133).

  6. Supplementary Figure 6: Association between polygenic score and YOE, in individuals carrying 0 or ≥1 URVs or pathogenic CNVs (27 KB)

    Only Swedish participants are included (N=10,651).

  7. Supplementary Figure 7: Q-Q plots for the gene-based burden test for association between disruptive and damaging variants and YOE. (151 KB)

    Each point is a gene. Results are from meta-analysis across studies (N=14,133). In the upper panels we included only URVs, in the lower panels we included variants with minor allele frequency < 0.05%.

PDF files

  1. Supplementary Text and Figures (1,131 KB)

    Supplementary Figures 1–7

  2. Supplementary Methods Checklist (406 KB)

Excel files

  1. Supplementary Table 1 (10,247 KB)

    Educational attainment coding for each study

  2. Supplementary Table 2 (10,316 KB)

    Distribution of YOE by study, year of birth and sex

  3. Supplementary Table 3 (10,347 KB)

    Number and distribution of URVs by study

  4. Supplementary Table 4 (13,697 KB)

    Large pathogenic CNVs included in the study and number of carriers

Additional data