Skip to main content

Thank you for visiting 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.

Genetic and functional insights into the fractal structure of the heart


The inner surfaces of the human heart are covered by a complex network of muscular strands that is thought to be a remnant of embryonic development1,2. The function of these trabeculae in adults and their genetic architecture are unknown. Here we performed a genome-wide association study to investigate image-derived phenotypes of trabeculae using the fractal analysis of trabecular morphology in 18,096 participants of the UK Biobank. We identified 16 significant loci that contain genes associated with haemodynamic phenotypes and regulation of cytoskeletal arborization3,4. Using biomechanical simulations and observational data from human participants, we demonstrate that trabecular morphology is an important determinant of cardiac performance. Through genetic association studies with cardiac disease phenotypes and Mendelian randomization, we find a causal relationship between trabecular morphology and risk of cardiovascular disease. These findings suggest a previously unknown role for myocardial trabeculae in the function of the adult heart, identify conserved pathways that regulate structural complexity and reveal the influence of the myocardial trabeculae on susceptibility to cardiovascular disease.

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

Relevant articles

Open Access articles citing this article.

Access options

Rent or buy this article

Get just this article for as long as you need it


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

Fig. 1: Trabeculation phenotypes and covariates.
Fig. 2: Genetic associations of left ventricular trabeculation.
Fig. 3: Knockout of mtss1 leads to a reduction of cardiac trabeculation in medaka.
Fig. 4: Relationship between trabecular complexity and cardiac function and disease.

Data availability

The genetic and phenotypic UK Biobank data presented in this study are available to any bona fide researcher upon application to UK Biobank ( The GWAS summary level data used in this study are publicly available: per-slice and meta-analysis summary statistics of fractal dimension can be accessed through the GWAS catalogue (; accession numbers GCST90000287–GCST90000296), LD score regression traits are available at and heart failure summary statistics can be found at Figs. 1, 3 and 4 contain raw data that are provided as source data unless prior application to UK Biobank is required. Source data are provided with this paper.

Code availability

The analysis code is freely available on GitHub (


  1. Sedmera, D. & McQuinn, T. Embryogenesis of the heart muscle. Heart Fail. Clin. 4, 235–245 (2008).

    PubMed  PubMed Central  Google Scholar 

  2. Sizarov, A. et al. Formation of the building plan of the human heart: morphogenesis, growth, and differentiation. Circulation 123, 1125–1135 (2011).

    PubMed  Google Scholar 

  3. Kawabata Galbraith, K. et al. MTSS1 regulation of actin-nucleating formin DAAM1 in dendritic filopodia determines final dendritic configuration of Purkinje cells. Cell Rep. 24, 95–106 (2018).

    CAS  PubMed  Google Scholar 

  4. Praschberger, R. et al. Mutations in Membrin/GOSR2 reveal stringent secretory pathway demands of dendritic growth and synaptic integrity. Cell Rep. 21, 97–109 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  5. Chen, X. et al. Knockout of SRC-1 and SRC-3 in mice decreases cardiomyocyte proliferation and causes a noncompaction cardiomyopathy phenotype. Int. J. Biol. Sci. 11, 1056–1072 (2015).

    CAS  PubMed  PubMed Central  Google Scholar 

  6. Luxán, G., D’Amato, G. & de la Pompa, J. L. Intercellular Signaling in Cardiac Development and Disease: The NOTCH pathway 103–114 (Springer Japan, 2016).

  7. Han, P. et al. Coordinating cardiomyocyte interactions to direct ventricular chamber morphogenesis. Nature 534, 700–704 (2016).

    CAS  PubMed  PubMed Central  ADS  Google Scholar 

  8. Captur, G. et al. Morphogenesis of myocardial trabeculae in the mouse embryo. J. Anat. 229, 314–325 (2016).

    CAS  PubMed  PubMed Central  Google Scholar 

  9. Miquerol, L. et al. Biphasic development of the mammalian ventricular conduction system. Circ. Res. 107, 153–161 (2010).

    CAS  PubMed  Google Scholar 

  10. del Monte-Nieto, G. et al. Control of cardiac jelly dynamics by NOTCH1 and NRG1 defines the building plan for trabeculation. Nature 557, 439–445 (2018).

    PubMed  ADS  Google Scholar 

  11. van Weerd, J. H. & Christoffels, V. M. The formation and function of the cardiac conduction system. Development 143, 197–210 (2016).

    PubMed  Google Scholar 

  12. Vedula, V., Seo, J.-H., Lardo, A. C. & Mittal, R. Effect of trabeculae and papillary muscles on the hemodynamics of the left ventricle. Theor. Comput. Fluid Dyn. 30, 3–21 (2016).

    Google Scholar 

  13. Sacco, F. et al. Left ventricular trabeculations decrease the wall shear stress and increase the intra-ventricular pressure drop in CFD simulations. Front. Physiol. 9, 458 (2018).

    PubMed  PubMed Central  Google Scholar 

  14. Paun, B., Bijnens, B. & Butakoff, C. Relationship between the left ventricular size and the amount of trabeculations. Int. J. Numer. Methods Biomed. Eng. 34, e2939 (2018).

    MathSciNet  Google Scholar 

  15. van Waning, J. I. et al. Genetics, clinical features, and long-term outcome of noncompaction cardiomyopathy. J. Am. Coll. Cardiol. 71, 711–722 (2018).

    PubMed  Google Scholar 

  16. Bycroft, C. et al. The UK Biobank resource with deep phenotyping and genomic data. Nature 562, 203–209 (2018).

    CAS  PubMed  PubMed Central  ADS  Google Scholar 

  17. Petersen, S. E. et al. UK Biobank’s cardiovascular magnetic resonance protocol. J. Cardiovasc. Magn. Reson. 18, 8 (2015).

    Google Scholar 

  18. Tarroni, G. et al. Learning-based quality control for cardiac MR images. IEEE Trans. Med. Imaging 38, 1127–1138 (2019).

    PubMed  Google Scholar 

  19. Bai, W. et al. A bi-ventricular cardiac atlas built from 1000+ high resolution MR images of healthy subjects and an analysis of shape and motion. Med. Image Anal. 26, 133–145 (2015).

    PubMed  ADS  Google Scholar 

  20. Shah, S. et al. Genome-wide association and Mendelian randomisation analysis provide insights into the pathogenesis of heart failure. Nat. Commun. 11, 163 (2020).

    CAS  PubMed  PubMed Central  ADS  Google Scholar 

  21. Bai, W. et al. Automated cardiovascular magnetic resonance image analysis with fully convolutional networks. J. Cardiovasc. Magn. Reson. 20, 65 (2018).

    PubMed  PubMed Central  Google Scholar 

  22. Captur, G. et al. Fractal analysis of myocardial trabeculations in 2547 study participants: multi-ethnic study of atherosclerosis. Radiology 277, 707–715 (2015).

    PubMed  PubMed Central  Google Scholar 

  23. MacArthur, J. et al. The new NHGRI-EBI Catalog of published genome-wide association studies (GWAS Catalog). Nucleic Acids Res. 45, D896–D901 (2017).

    CAS  PubMed  Google Scholar 

  24. Sudlow, C. et al. UK Biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 12, e1001779 (2015).

    PubMed  PubMed Central  Google Scholar 

  25. Zerbino, D. R., Wilder, S. P., Johnson, N., Juettemann, T. & Flicek, P. R. The ensembl regulatory build. Genome Biol. 16, 56 (2015).

    PubMed  PubMed Central  Google Scholar 

  26. Burkhoff, D., Mirsky, I. & Suga, H. Assessment of systolic and diastolic ventricular properties via pressure–volume analysis: a guide for clinical, translational, and basic researchers. Am. J. Physiol. Heart Circ. Physiol. 289, H501–H512 (2005).

    CAS  PubMed  Google Scholar 

  27. Seemann, F. et al. Noninvasive quantification of pressure-volume loops from brachial pressure and cardiovascular magnetic resonance. Circ. Cardiovasc. Imaging 12, e008493 (2019).

    PubMed  Google Scholar 

  28. Zheng, J. et al. LD Hub: a centralized database and web interface to perform LD score regression that maximizes the potential of summary level GWAS data for SNP heritability and genetic correlation analysis. Bioinformatics 33, 272–279 (2017).

    CAS  PubMed  Google Scholar 

  29. Hemani, G. et al. The MR-Base platform supports systematic causal inference across the human phenome. eLife 7, e34408 (2018).

    PubMed  PubMed Central  Google Scholar 

  30. Smith, G. D. et al. Clustered environments and randomized genes: a fundamental distinction between conventional and genetic epidemiology. PLoS Med. 4, e352 (2007).

    PubMed  PubMed Central  Google Scholar 

  31. Hemani, G., Tilling, K. & Davey Smith, G. Orienting the causal relationship between imprecisely measured traits using GWAS summary data. PLoS Genet. 13, e1007081 (2017).

    PubMed  PubMed Central  Google Scholar 

  32. Keele, K. Leonardo Da Vinci’s Elements of the Science of Man (Academic, 2014).

  33. Jensen, B., Wang, T., Christoffels, V. M. & Moorman, A. F. Evolution and development of the building plan of the vertebrate heart. Biochim. Biophys. Acta 1833, 783–794 (2013).

    CAS  PubMed  Google Scholar 

  34. Brutsaert, D. L. Cardiac endothelial–myocardial signaling: its role in cardiac growth, contractile performance, and rhythmicity. Physiol. Rev. 83, 59–115 (2003).

    CAS  PubMed  Google Scholar 

  35. Morley, M. P. et al. Cardioprotective effects of MTSS1 enhancer variants. Circulation 139, 2073–2076 (2019).

    PubMed  PubMed Central  Google Scholar 

  36. Wild, P. S. et al. Large-scale genome-wide analysis identifies genetic variants associated with cardiac structure and function. J. Clin. Invest. 127, 1798–1812 (2017).

    PubMed  PubMed Central  Google Scholar 

  37. Aung, N. et al. Genome-wide analysis of left ventricular image-derived phenotypes identifies fourteen loci associated with cardiac morphogenesis and heart failure development. Circulation 140, 1318–1330 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  38. Kenkel, N. C. & Walker, D. J. Fractals in the biological sciences. Coenoses 11, 77–100 (1996).

    Google Scholar 

  39. Olejníčková, V., Šaňková, B., Sedmera, D. & Janáček, J. Trabecular architecture determines impulse propagation through the early embryonic mouse heart. Front. Physiol. 9, 1876 (2019).

    PubMed  PubMed Central  Google Scholar 

  40. Ingles, J. et al. Evaluating the clinical validity of hypertrophic cardiomyopathy genes. Circ. Genom. Precis. Med. 12, e002460 (2019).

    CAS  PubMed  PubMed Central  Google Scholar 

  41. Schafer, S. et al. Titin-truncating variants affect heart function in disease cohorts and the general population. Nat. Genet. 49, 46–53 (2017).

    CAS  PubMed  Google Scholar 

  42. Miszalski-Jamka, K. et al. Novel genetic triggers and genotype–phenotype correlations in patients with left ventricular noncompaction. Circ. Cardiovasc. Genet. 10, e001763 (2017).

    CAS  PubMed  PubMed Central  Google Scholar 

  43. Tayal, U., Prasad, S. & Cook, S. A. Genetics and genomics of dilated cardiomyopathy and systolic heart failure. Genome Med. 9, 20 (2017).

    PubMed  PubMed Central  Google Scholar 

  44. Munafò, M. R. & Davey Smith, G. Robust research needs many lines of evidence. Nature 553, 399–401 (2018).

    PubMed  ADS  Google Scholar 

Download references


The research was supported by the UK Medical Research Council (MC-A651-53301); British Heart Foundation (NH/17/1/32725, RG/19/6/34387, RE/18/4/34215); Wellcome Trust (107469/Z/15/Z); National Institute of Environmental Health Sciences (R01 ES029917-02); Heidelberg University; the Simons Center for Quantitative Biology at Cold Spring Harbor Laboratory; and the National Institute for Health Research Biomedical Research Centre based at Imperial College Healthcare NHS Trust and Imperial College London. P.M.M. has also received personal and research support from the Edmond J. Safra Foundation and Lily Safra, an NIHR Senior Investigator’s Award, the Medical Research Council and the UK Dementia Research Institute. R.T.L. is supported by a UKRI Health Data Research Rutherford Fellowship (MR/S003754/1). A.H. is supported by a BHF PhD studentship. J.G. is supported by a Research Center for Molecular Medicine (HRCMM) Career Development Fellowship, the MD/PhD program of the Medical Faculty Heidelberg, the Deutsche Herzstiftung (S/02/17) and by an Add-On Fellowship for Interdisciplinary Science of the Joachim Herzstiftung. Research funding for cohorts used in Mendelian Randomization: NIHR Cardiovascular Biomedical Research Unit of Royal Brompton and Harefield NHS Foundation Trust (DCM cohort). Funding information for HERMES-participating studies is described elsewhere20. Data aggregation and downstream bioinformatics were funded through grants from the MRC Proximity to Discovery scheme, the NIHR UCLH Biomedical Research Centre and the EU/EFPIA Innovative Medicines Initiative 2 Joint Undertaking BigData@Heart grant no. 116074. We thank H. Suzuki for his work on preprocessing the image data, R. Fumero for advice on the finite element modelling, V. Uhlmann for advice on radial image registration, L. Schertel and C. Baader for sgRNA production, members of the Wittbrodt Laboratory for discussion and support and B. Statton and M. Thanaj for assisting with data preprocessing.

Author information

Authors and Affiliations



H.V.M. and T.J.W.D. performed the formal analysis and co-wrote the manuscript; M.S. and M.L.C. performed the in silico modelling; T.J.W.D. and A.d.M. collected and analysed image data; R.T.L., A.H., J.S.W. and S.K.P. collected and analysed the clinical data; W.B., P.T., J.C. and D.R. developed the computational phenotyping; J.G., T.T. and J.W. detailed the experimental strategy for the medaka validation; J.G. and T.T. designed and performed CRISPR–Cas9 knockout experiments and conducted phenotypic analysis under the guidance of J.W.; J.G. acquired light sheet microscopy recordings, and analysed and plotted the medaka knockout and imaging data. P.M.M., E.B., S.A.C. and D.P.O. provided interpretation of the results; E.B., S.A.C. and D.P.O. conceived the study, managed the project and revised the manuscript. All authors reviewed the final manuscript.

Corresponding authors

Correspondence to Hannah V. Meyer or Declan P. O’Regan.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature thanks Tim Leiner, Christopher Newton-Cheh and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data figures and tables

Extended Data Fig. 1 The ethnicities of participants in the discovery and replication cohorts.

a, b, Principal components 1 and 2 of the principal component analysis of the combined genotypes of the HapMap III datasets (n = 1,184 individuals) and the UK Biobank discovery cohort (a; n = 19,262 individuals; 159,243 independent genetic variants) or the UK Digital Heart study (b; n = 2,985 individuals; 149,707 independent genetic variants). Data from the UK Biobank (a) or UK Digital Heart cohort (b) are depicted in blue, HapMap individuals are coloured by their ethnicity. Cohort individuals within 1.5 standard deviations distance from the centre of the European HapMap individuals (grey) are selected for further analyses.

Extended Data Fig. 2 Fractal dimension phenotypes.

a, The distribution of CMR image slices for which the fractal dimensions were measured in the n = 19,761 UK Biobank individuals. Missing fractal dimension measurements per slice can arise because a slice was not measured (NA), or the estimation of the fractal dimension failed owing to poor image quality or the estimated fractal dimension failing quality control (NaN). b, The correlation between fractal dimension summary measures derived from the observed fractal dimension slice measurements and interpolated fractal dimension measurements per sample. Interpolated fractal dimension measurements per sample were derived by using a Gaussian kernel local fit to different numbers of slice templates, allowing for direct slice comparisons across individuals. Different numbers of slices for interpolation were tested (rows; 7–12 slices). Columns show different summary measures: the mean fractal dimension across all measured slices (MeanGlobalFD) or mean fractal dimension per slice region (MeanBasalFD, MeanMidFD, MeanApicalFD) were analysed. Linear model of measured~interpolated (r2) values of the summary measures between measured and interpolated fractal dimensions in the UK Biobank (n = 19,761 individuals).

Extended Data Fig. 3 Acquisition and processing of phenotypes.

a, Fractal dimension analysis on cardiac computed tomography images. The fractal dimension was calculated using the same method as for CMR, but with manual regions of interest, in a set of gated cardiac computed tomography images. Top, analogous processing as described in Fig. 1c, using edge detection of the trabeculae and subsequent box counting across a range of sizes. Bottom, analogous to Fig. 2a, box plots of the fractal dimension measurements for 20 individuals per slice, colour-coded by cardiac region. The lower and upper hinges in the box plot correspond to the 25th and 75th percentiles (IQR), the horizontal line in the box plot indicates the median. The lower and upper whiskers extend from the hinge to the smallest and largest values that are no further than 1.5× the IQR. b, Myocardial strain. Global longitudinal Lagrangian strain at each cardiac phase for all UK Biobank participants with CMR imaging (n = 26,893). Individual data points shown with a smoothed mean and density contours. c, Principal component analysis of fractal dimension measurements across all 9 slices in the 18,096 individuals of the UK Biobank discovery cohort. Proportion of variance explained for each principal component (left). Biplot of the first and second (middle) or third and fourth (right) principal components of each individual (grey points). The corresponding loadings for the fractal dimension of slices 1–9 are shown as vectors. d, Genotype, fractal dimension and trabeculation outlines for rs35006907. Representative, registered, trabecular outlines at slice 5 are shown for the median fractal dimension of individuals with the homozygous major (blue), heterozygous (pink) and homozygous minor (green) genotypes of rs35006907. e, Pearson correlation of global fractal dimension and QRS duration (n = 18,096 individuals). QRS duration phenotype from UKB ID: qrs_duration_f12340_2_0. The Pearson correlation coefficient is indicated in the top right corner.

Extended Data Fig. 4 Per-slice fractal dimension GWAS and inclusion of additional covariates.

a, b, Manhattan plots (a) and quantile–quantile plots (b) of the independently conducted, nine univariate GWAS on the per-slice fractal dimension measurements for 18,096 UK Biobank individuals passing quality control. a, In the Manhattan plots, the P values (derived from linear association t-statistics) were multiplied by the effective number of independent phenotypic tests Teff = 6.6 and min(Padjust, 1) reported. In the quantile–quantile plots, the unadjusted P values are plotted against equally spaced values in the range of 0–1 of the same sample size (expected P values). The diagonal line starts at the origin and has a slope of one. The genomic control λ values for each quantile–quantile plot are: 1.0557, 1.0436, 1.0496, 1.0557, 1.0649, 1.0679, 1.0679, 1.0618 and 1.0436. λ values were generated with linkage disequilibrium score regression, for details see Supplementary Table 1. c, d, Manhattan plots based on the meta-analysis GWAS (sample size, n = 18,096 individuals) with end-diastolic volume of the left ventricle (c) or myocardial strain (d) as co-variate. e, Manhattan plot based on the meta-analysis GWAS (this panel is the same as Fig. 2a and is shown for comparison). Other co-variates and analysis parameters (as described in the Supplementary Methods) were kept the same in all analyses (ae). P values are meta-analysis P values, not adjusted for multiple testing derived from the transformation of the univariate signed t-statistics (associations on 14,134,301 genetic variants at 16 independent loci from 18,096 individuals) and χ2 distribution with 9 degrees of freedom. In a and ce, the horizontal grey line is drawn at the level of genome-wide significance: P = 5 × 10−8.

Extended Data Fig. 5 GWAS effect size estimates and replication.

a, Effect size distribution of loci with genetic variant associations of Padjust = 5 × 10−8 in any univariate per-slice fractal dimension GWAS (sample size, n = 18,096 individuals). P were values derived from linear association t-statistics. Distributions are shown for each locus (indicated by chromosomal position and lead genetic variant in the subplot title) across all slices and effect sizes, colour-coded by P value of the association. Variants with no Padjust < 5 × 10−8 in the univariate per-slice fractal dimension GWAS (all blue) were discovered in the multi-trait meta-analyses. b, c, Effect size estimate concordance in discovery and replication cohorts. For each of the nine univariate, per-slice fractal dimension GWAS, the effect size estimates of the genetic variants with the smallest P value for each of the independent loci in the discovery cohort (n = 18,096 individuals) were selected. For some variants, associations that passed the GWAS threshold of Padjust < 5 × 10−8 were discovered in more than one of the nine univariate GWAS fractal dimension slices; for these variants all effect size estimates were selected. Estimates were plotted against the corresponding slice-variant associations in the replication GWAS (b, UK Biobank replication, n = 6,356 individuals; c, UK Digital Heart cohort, n = 1,029 individuals). Non-concordant estimate pairs are depicted in light grey. Effect size estimates that passed the Bonferroni-adjusted validation P-value threshold of P < 0.05/16 = 0.003 are depicted as triangles. r2 for linear model of \({\hat{\beta }}_{{\rm{discovery}}} \sim {\hat{\beta }}_{{\rm{replication}}}\).

Extended Data Fig. 6 Annotation of trabeculation associated loci.

a, Gene expression of GTEx-associated genes and tissues. Gene expression is shown as log10-transformed transcripts per million (TPM) for genes for which the expression is associated with trabeculation loci (using GTEx look-up; Supplementary Data 3). Gene expression values and tissues were downloaded from by querying: gene name AND tissue AND species; that is, GTEx gene AND heart component AND Homo sapiens. Light grey tiles indicate gene expression values for the gene–tissue pair that are not available. b, Enrichment of trabeculation-associated variants in DNase I hypersensitive sites for all available tissues in GARFIELD. GARFIELD was used to compute the functional enrichment (odds ratio (OR)) of genetic variants associated with the trabeculation phenotypes (GWAS: n = 18,096 individuals, P values derived from linear association t-statistics) at P < 10−6 for open chromatin regions. The results across all available studies per tissue are shown as box plots. Lower and upper hinges, 25th and 75th percentiles (IQR); horizontal line, median; lower and upper whiskers extend from the hinge to the smallest and largest value no further than 1.5× the IQR.

Extended Data Fig. 7 Biomechanical model, genetic correlation and disease associations.

a, Left ventricular pressure–volume loops from finite-element modelling across a range of atrial pressures. Solid and dashed lines indicate smooth and trabeculated ventricles, respectively. b, Mid-short-axis cross-sections of the finite element model of the left ventricle, looking towards the apex, at different trabecular complexities. c, The ventricular model was in series with pre-load (red) and after-load (blue) circuits that define the left atrial pressure (PLA), right atrial pressure (PRA), inflow resistance (R1), aortic resistance (R2), peripheral resistance (R3) and vascular capacitance (c). Initial parameters calibrated to approximate observations from UK Biobank data; the reference model was a trabeculated left ventricle with a PLA of 5 mm Hg. d, Fractal dimension association P values (depicted on −log10 scale, uncorrected for multiple comparisons; estimated by transformation of univariate GWAS signed t-statistics with χ2 distributions with 9 degrees of freedom; univariate GWAS with n = 18,096 individuals) for the GOSR2 locus on chromosome 17; variants associated with mixed aetiology heart failure (n = 47,309 cases, n = 930,014 controls) and DCM (n = 510 cases, n = 1,136 controls) are highlighted in purple. eg, Summary statistics of basal, mid and apical trabeculation GWAS were analysed for genetic correlation with all available summary statistics on LD Hub. e, Additive heritability estimates h2 for regional summary statistics based on 1,208,036 genetic variants. f, Association P values of heart and cardiovascular phenotypes with corresponding estimate of genetic correlation (encoded by size). g, Genetic correlation P values of all available LD Hub traits (x axis, ordered by category) with trabeculation GWAS results (based on linkage disequilibrium score regression correlation of 1,208,036 genetic variants) by region summarized in LD Hub categories (colour-coded). Heart and cardiovascular phenotypes (y axis in f and Supplementary Table 13) are depicted in magenta. g, P values are derived from cross-trait correlation analysis and a block jackknife approach for estimation of the standard error of the genetic correlation (Supplementary Table 13); P values are shown on −log10 scale and are uncorrected for multiple comparisons.

Extended Data Fig. 8 Mendelian randomization analysis of the effect of trabeculation on heart failure and DCM.

a, Mendelian randomization on heart failure. The effect size estimates for heart failure are based on n = 47,309 cases and n = 930,014 controls in the HERMES study20. b, Mendelian randomization on DCM. The effect size estimates for DCM are based on n = 1,134 cases and n = 510 controls. The fractal dimension effect size estimates from univariate GWAS results on n = 18,096 samples are shown. Scatter plots (top left) depict the genetic variant-exposure effect versus the genetic variant-outcome effect. Centre values show effect size estimates on fractal dimension and DCM, error bars show the standard error of the association test (t-statistic for fractal dimension, logistic regression for heart failure and DCM). Forest plots (top right) show the contribution of each genetic variant to the overall estimate (black; estimated by Wald ratio) and combined as a single genetic instrument (purple; estimated by the indicated method) for the four tested Mendelian randomization methods (these panels are the same as Fig. 4c,d and are shown for convenient comparison). Funnel plots (bottom left) depict the instrument strength against the causal effect of each instrument as a single instrumental variable. Vertical lines indicate the average estimated effect for the tested Mendelian randomization methods. Strong instruments are close to the estimated average effect, while weak instruments spread evenly on both sides. Leave-one-out plots (bottom right) show the results of Mendelian randomization analysis (inverse-variance weighted only), in which each genetic variant is sequentially excluded. This analysis can indicate whether there are any single variants that drive the Mendelian randomization results. In the right panels, centre values mark effect size point estimates, error bars show the 95% confidence intervals.

Extended Data Table 1 Participant characteristics
Extended Data Table 2 Annotations of trabeculation-associated loci

Supplementary information

Supplementary Information

This file contains Supplementary Notes on Mendelian Randomisation, Supplementary Methods, Supplementary References, Supplementary Tables 1-18 and Supplementary Figure 1.

Reporting Summary

Supplementary Data

This file contains Supplementary Data files 1-6 and a guide.

Video 1

: Control-injected medaka embryo at 4 days post fertilisation. Representative video of one control embryo (out of n=87 embryos). Corresponds to Figure 3c, control embryo. LSM was carried out on a 16x multiview selective plane illumination microscope (MuVi-SPIM). Entire heart volumes were acquired with 488 nm illumination, a 525/50 nm bandpass filter, and a z step size of 2 μm. Surface rendering was performed with UCSF Chimera (1.11).

Video 2

: Moderately affected mtss1 KO embryo at 4 days post fertilisation. Representative video of one mtss1 KO embryo (out of n=13 embryos). Corresponds to Figure 3c, mtss1 KO. It shows abnormal cardiac morphology and reduced amount of intracardiac blood volume. LSM was carried out on a 16x multiview selective plane illumination microscope (MuVi-SPIM). Entire heart volumes were acquired with 488 nm illumination, a 525/50 nm bandpass filter, and a z step size of 2 μm. Surface rendering was performed with UCSF Chimera (1.11).

Source data

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Meyer, H.V., Dawes, T.J.W., Serrani, M. et al. Genetic and functional insights into the fractal structure of the heart. Nature 584, 589–594 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:

This article is cited by


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing