Sibling DNA study estimated heritability for 14 traits across diverse ancestries
Differences in DNA sharing among approximately 500,000 sibling pairs yielded estimates of the genetic contribution to variation in 14 traits, including behaviour and common diseases. The study in Nature Human Behaviour used data from 23andMe Research Institute participants. Estimates were 0.70 ± 0.03 for height and 0.49 ± 0.03 for body mass index, a measure relating weight to height. Results were consistent across genetic ancestry groups. Shared environmental effects may have inflated estimates from comparisons between families for behaviour and type 2 diabetes risk, the authors found.
Heritability measures how much of the variation in a trait between individuals is explained by genetic factors. The study's estimates were lower than those from twin studies, but larger than the variation accounted for by genome-wide association studies, which search across DNA for statistical links to traits.
The researchers analysed up to 547,126 full sibling pairs across European, East Asian, African American and Latino groups. Approximately 40% of participants were male, and their average age was approximately 50 years. The authors described the African American and Latino categories as ethnicity groups because participants typically had a mixture of genetic ancestries.
Siblings inherit different combinations of their parents' DNA. The researchers compared differences in how much DNA siblings shared with differences in their traits. This approach is robust to most factors that can distort estimates, including shared environments and genetic differences between populations. Its limitation is statistical power: precise estimates require hundreds of thousands of sibling pairs.
Within-family heritability estimates were statistically significant for smoking initiation (0.34 ± 0.05) and alcohol consumption (0.18 ± 0.04). They were also significant for number of children (0.27 ± 0.11) and the personality measure ‘talk versus listen’ (0.48 ± 0.13). Estimates for the two common diseases were 0.43 ± 0.06 for type 2 diabetes and 0.34 ± 0.06 for asthma.
The study provides within-family estimates across diverse ancestries for traits and diseases that previously lacked them. By reducing distortion from shared environments and population differences, the approach helps researchers assess how much genetic variation contributes to differences between people and where comparisons between families may overstate that contribution.
Written by the Genomes desk from the primary source cited and linked above and checked against it. Research use only; not medical advice. Corrections: [email protected].