Quantification and dissection of the nature-nurture question using genomic and family data

The nature-nurture question in human genetics relies on precise and unbiased estimation of genetic variance. Traditionally, estimation of heritability was based upon pedigree information but we introduced a method to estimate it within families by exploiting genome-wide marker data (Visscher et al. 2006, PLOS Genetics; Visscher et al. 2007, Am J Hum Genet; Hemani et al., 2013, Am J Hum Genet). The largest application of these methods was on ~20,000 sibling pairs (Hemani et al. 2013) and even then the standard errors of the estimates were large. A generalization of the within-family method was proposed recently, reporting smaller estimates of heritability when compared to twin studies, albeit also with large standard errors (Young et al. 2018, Nature Genetics).

We have also developed methods to estimate genetic variation from genome-wide association study (GWAS) data that exclude close relatives (e.g. Yang et al. 2010, Nature Genetics; Yang et al. 2011, Am J Hum Genet). The contrast between these estimates and those from pedigree studies are informative with respect to the genetic architecture of complex traits (including disease) and the nature-nurture quantification.

We now propose to join those methodologies into powerful analyses to jointly estimate between and within-family genetic variance from the largest sample of sibling pairs and family data to date.