Sequential recruitment of study participants may inflate genetic heritability estimates

Abstract

After the success of genome-wide association studies to uncover complex trait loci, attempts to explain the remaining genetic heritability (h ) are mainly focused on unraveling rare variant associations and gene-gene or gene-environment interactions. Little attention is paid to the possibility that h estimates are inflated as a consequence of the epidemiological study design. We studied the time series of 54 biochemical traits in 4373 individuals from the Cooperative Health Research In South Tyrol (CHRIS) study, a pedigree-based study enrolling ten participants/day over several years, with close relatives preferentially invited within the same day. We observed distributional changes of measured traits over time. We hypothesized that the combination of such changes with the pedigree structure might generate a shared-environment component with consequent h inflation. We performed variance components (VC) h estimation for all traits after accounting for the enrollment period in a linear mixed model (two-stage approach). Accounting for the enrollment period caused a median h reduction of 4%. For 9 traits, the reduction was of>20%. Results were confirmed by a Bayesian Markov chain Monte Carlo analysis with all VCs included at the same time (one-stage approach). The electrolytes were the traits most affected by the enrollment period. The h inflation was independent of the h magnitude, laboratory protocol changes, and length of the enrollment period. The enrollment process may induce shared-environment effects even under very stringent and standardized operating procedures, causing h inflation. Including the day of participation as a random effect is a sensitive way to avoid overestimation.

Effects of smoking status, history and intensity on heart rate variability in the general population: The CHRIS study

“…The Cooperative Health Research In South Tyrol (CHRIS) study is a population-based study being carried out in South Tyrol (Italy) to investigate the basis of chronic conditions associated with human ageing. The study has been extensively described elsewhere [21, 22]. All participants underwent blood drawing, urine collection, anthropometry analysis, and clinical assessments in the early morning following overnight fasting.…”

“…Detailed medical history was reconstructed through both interviewer- and self-administered questionnaires. Participants were asked to report genealogical information about parents and grandparents, allowing the reconstruction of up-to-five generation pedigrees [22].…”

“…After excluding 20 pregnant women and subjects with no information on either HRV or smoking, the available sample size for analysis was of 4751 subjects. Given the relatedness structure of the CHRIS study sample [22], we fitted linear mixed regression models (LMMs) to explore the relationship between HRV indicators and functions of sex, age, and smoking constructs, also allowing for non-linearity and moderating effects. The LMMs’ covariance structure was based on a between-subjects’ relatedness matrix, where pairwise kinship coefficients were estimated from the pedigree using the ‘kinship2’ R package.…”

The UMOD Locus: Insights into the Pathogenesis and Prognosis of Kidney Disease

“…The fact that all loci reported so far only explain approximately 4% of the total variance of eGFRcrea, 31 whereas the genetic heritability of the trait is between 30% and 50% 88 means that most of the genetic heritability of eGFRcrea remains unexplained. That pessimistic view could be tempered by growing literature showing that genetic heritability estimates may be inflated 89 : issues include the design of epidemiologic studies 90 and unmodeled genetic and environmental components. 91,92 Strategies to expand large genomic studies related to renal function and CKD include larger biobanks with hundreds of thousands of genotyped individuals (e.g., the UK Biobank).…”

Comparative assessment of different familial aggregation methods in the context of large and unstructured pedigrees

“…After removing singletons, the dataset includes 4373 phenotyped individuals who reported two generations of ancestors resulting in 186 pedigrees with a total of 9024 individuals, out of which 4651 remain unphenotyped, but are used for establishing kinship (Noce et al , 2017). One exceptionally large family (termed ‘XL’) includes 3676 phenotyped and 3997 unphenotyped individuals.…”