Fast phylogenetic generalised linear mixed-effects modelling using the glmmTMB R package
Williams, C.; McGillycuddy, M.; Drobniak, S. M.; Bolker, B. M.; Warton, D. I.; Nakagawa, S.
Show abstract
Phylogenetic generalised linear mixed models (PGLMMs) help ecologists to distinguish ecological drivers from other processes shaping evolutionary patterns, yet existing implementations are often limited in distributional scope or computational speed. We compare five R packages for fitting PGLMMs and highlight the new covariance structure propto in the general-purpose GLMM package glmmTMB. Simulations show that glmmTMB fits PGLMMs faster overall than brms, MCMCglmm, INLA, and phyr, while producing similar model estimates. We present the first practical application of glmmTMB for fitting phylogenetic random effects using likelihood-based models that accommodate repeated measures, demonstrated through case studies of evolutionary trait data. By improving both speed and flexibility, glmmTMB broadens access to PGLMM and supports deeper insights into trait evolution and diversification.
Matching journals
The top 4 journals account for 50% of the predicted probability mass.