Implications of multilocus inheritance for gene-disease association studies

Area of Science:

• Genetics
• Biostatistics

Background:

• Candidate gene association studies are crucial for understanding disease genetics.
• Multilocus inheritance and background heritability significantly influence study power and parameter estimation.
• Existing methods may not fully account for the complexities of multilocus inheritance in diverse family structures.

Purpose of the Study:

• To evaluate the impact of multilocus inheritance on candidate gene association studies.
• To assess how background heritability affects study power and genotype relative risk (GRR) estimates.
• To compare the utility of multiplex families versus singletons and different control designs.

Main Methods:

• Theoretical analysis of statistical power and genotype relative risk estimation.
• Inclusion of both case-control and family-based (Transmission Disequilibrium Test - TDT) control designs.
• Consideration of varying levels of background heritability (H) and family structures (multiplex sibships, singletons, MZ twins).

Main Results:

• Study power is diminished in multiplex families when background heritability is high, favoring their use in low heritability disorders.
• Genotype relative risk (GRR) estimates are distorted in multiplex sibships; inflated in case-control studies and deflated in TDT studies, with effects moderated by heritability.
• Differences in genotype frequencies between concordant and discordant MZ twin pairs are larger for low heritability traits.

Conclusions:

• The choice of family structure and control design in association studies is critically dependent on the background heritability of the disease.
• Multiplex families offer advantages for low heritability disorders, while singleton designs may be more robust for high heritability conditions.
• Accurate estimation of genetic effects requires careful consideration of multilocus inheritance patterns and background heritability.