We are delighted to announce that Professor Matthew Keller, Director of the institute for Behavioral Genetics, University of Colorado Boulder, will give a guest lecture at the Department of Psychology. Professor Keller is an internationally recognized researcher in behavior genetics and psychiatric genetics, with a particular focus on how genetic and environmental factors interact in shaping mental health. His work has significantly advanced our understanding of complex traits, depression, and evolutionary perspectives on human psychology. Over the years he has developed structural equation models to model how parents affect their children through the environment and the impact of partner selection on intergenerational transmission.
Abstract
Structural equation modeling with polygenic scores (SEM-PGS) extends the approach of Kong et al. (2018), which used transmitted and non-transmitted haplotypic polygenic scores to estimate genetic nurture, by embedding it within a structural equation framework. This allows simultaneous estimation of additive genetic direct effects, vertical transmission (the causal influence of parental traits on offspring traits), genetic nurture, and different forms of assortative mating (phenotypic, social, or genetic) and their consequences for genetic and environmental variance. Recent developments generalize the model to a multivariate context, enabling decomposition of cross-trait pathways and making it one of the first family-based genetically informed frameworks capable of parsing within- and cross-trait influences. Because the method requires phased genotypes and phenotypes on parents and offspring, population-scale trio cohorts such as MoBa are uniquely valuable. I will present early MoBa results on spousal resemblance and intergenerational transmission for traits including educational attainment, height, and ADHD, showing both direct genetic contributions and continuing parental influences after accounting for shared genes. I will also discuss current limitations and plans to relax assumptions and broaden the model, underscoring SEM-PGS as a versatile next-generation framework for understanding how genes, environments, and social structures shape intergenerational transmission.