Genetic Analysis Workshop II: pedigree analysis of a binary trait without assuming an underlying liability

Area of Science:

• Biostatistics
• Statistical Genetics
• Quantitative Genetics

Background:

• Traditional models for binary trait concordance often assume an underlying latent liability threshold.
• Previous work established a model for twin pairs that avoids this latent liability assumption.
• Extending such models to complex pedigrees is crucial for broader applicability.

Purpose of the Study:

• To extend a concordance model, previously developed for twin pairs, to pedigrees of arbitrary structure.
• To evaluate the model's performance in recovering known structures from simulated data.
• To explore the implications of an assumption equivalent to multiplicative relative risks for low incidence rates.

Main Methods:

• The study employs a statistical modeling approach for binary trait concordance.
• The model is generalized from twin pairs to arbitrary pedigrees.
• An assumption is introduced that approximates multiplicative relative risks for low incidence rates.
• The model is validated using simulated workshop data with known structures.

Main Results:

• The generalized model successfully accommodates complex family structures.
• Application to simulated data demonstrates the model's ability to recover known underlying structures.
• The assumption of multiplicative relative risks is shown to be nearly equivalent for low incidence rates.

Conclusions:

• The developed model provides a flexible framework for analyzing binary trait concordance in extended pedigrees.
• This approach offers an alternative to latent liability models, particularly useful in genetic epidemiology.
• The findings support the utility of the model for dissecting familial aggregation of binary traits.