Enriched power of disease-concordant twin-case-only design in detecting interactions in genome-wide association studies

Abstract

Genetic interaction is a crucial issue in the understanding of functional pathways underlying complex diseases. However, detecting such interaction effects is challenging in terms of both methodology and statistical power. We address this issue by introducing a disease-concordant twin-case-only design, which applies to both monozygotic and dizygotic twins. To investigate the power, we conducted a computer simulation study by setting a series of parameter schemes with different minor allele frequencies and relative risks. Results from the simulation study reveals that the disease-concordant twin-caseonly design largely reduces sample size required for sufficient power compared to the ordinary case-only design for detecting gene-gene interaction using unrelated individuals. Sample sizes for dizygotic and monozygotic twins were roughly 1/2 and 1/4 of sample sizes in the ordinary case-only design. Since dizygotic twins are genetically similar as siblings, the enriched power for dizygotic twins also applies to affected siblings, which could help to largely extend the application of the powerful twin-case-only design. In summary, our simulation reveals high value of disease-concordant twins and siblings in efficiently detecting gene-by-gene interactions.

Genome-wide DNA methylation analysis of pulmonary function in middle and old-aged Chinese monozygotic twins

“…First, compared with other general case–control design, the sample size of our study was relatively small due to the difficulty of recruiting and identifying qualified MZ twin pairs. However, previous study has determined that the sample sizes of monozygotic twins just require roughly 1/4 of sample sizes in the ordinary case-only design to provide the sufficient power [ 65 ]. Second, the DNA sample was extracted from blood rather than the lung tissue.…”

RetroFun-RVS: a retrospective family-based framework for rare variant analysis incorporating functional annotations

“…One important feature of these approaches is that they focus on, or can be restricted to, only affected family members, when these are expected to contribute more information than unaffected subjects (Schaid et al, 2010). Affected-only designs have a long tradition in gene-gene or gene-environment interaction analysis and have been extended to family-based studies, requiring smaller sample sizes to reach equivalent power, compared to considering unrelated case-only individuals, which is an appealing feature in practice (W. Li et al, 2019). However, selecting individuals retrospectively (based on their phenotype) may lead to highly ascertained sampling schemes resulting in overestimated association measures.…”

“…Affected-only designs have a long tradition in gene-gene or gene-environment interaction analysis and have been extended to family-based studies, requiring smaller sample sizes to reach equivalent power, compared to considering unrelated case-only individuals, which is an appealing feature in practice 19 .…”

Exploring gene–gene interaction in family‐based data with an unsupervised machine learning method: EPISFA

“…Motivated from the idea that gene–gene interaction could be identified based on searching for the strength of SNP–SNP correlation (Frost, Amos, & Moore, 2016; Yang, Khoury, Sun, & Flanders, 1999), unsupervised machine learning methods can be adopted to search for G × G via learning relationship among these single‐nucleotide polymorphisms (SNPs). Tests derived from similar ideas have been assumed to have higher statistical power in exploring late‐onset diseases compared to case‐control methods and other family‐based tests including transmitted disequilibrium tests (Hu et al, 2014; Li et al, 2019). Also, the unsupervised machine learning has achieved satisfactory performance in computational efficiency while sustaining high statistical power, especially in a high‐dimensional dataset (Hassani, 2019).…”