A1448
Title: Functional mixed-effects models for function-valued traits in quantitative genetics
Authors: Yilin Chen - Kings College London (United Kingdom) [presenting]
Abstract: Function-valued traits, such as growth trajectories, can be described as a function of a continuous index. These traits are assessed for continuous genetic variation using a mixed-effects model to decompose phenotypic variation into additive genetic and environmental components. A functional genetic mixed-effects model is proposed, using functional principal component analysis to derive a data-driven basis that improves the approximation of the underlying data structure. The model separates phase and amplitude variability by warping functions to address curve misalignment. It extends the lme4 package to model genetic data in a functional framework, with between-level correlations induced by post-multiplying the design matrix by the Cholesky factor of the additive genetic relationship matrix. Simulation studies show that the model captures the genetic covariance function effectively. Future focus will be on jointly analyzing warping functions and curve data.
