Yi Li’s research mainly focuses on the integration of genomics and demographic studies on health and marriage. Social scientists are mostly interested in complex traits and behaviors such as health and education. Many of these outcomes are interactively influenced by social and biological factors. This highlights the importance of incorporating biological variables into the analysis of social science studies. The increasing availability of genetic, epigenetic, and biomarker data in social surveys enables researchers for the first time to address many new questions. Yi Li’s work attempts to enrich theoretical frameworks and to develop new empirical models for gene-environment interaction research.

An example is an article published in Journal of Marriage and Family. In this project the research question is whether marriage moderates the effect of 64 aggression-related genes on antisocial behavior. Previous studies show that marriage can inhibit antisocial behavior. Gene-environment studies show that the effects of social institutions on antisocial behavior depend on genetic factors.

However, existing gene-environment interaction studies almost exclusively focus on one or a few genetic variants. Because overall genetic influence on antisocial behavior is comprised of a huge number of genetic effects, it is important to consider more than a few genetic variants.

To do so, this project extends a recently developed mixed linear model implemented in the genome-wide complex traits analysis (GCTA) software (Nature Genetics 2010). This mixed linear model estimates narrow heritability—the proportion of variance in the phenotype explained by the genes. This project examines the gene-by-marriage interaction by comparing the proportion of variance in antisocial behavior explained by 64 genes among married and unmarried individuals. Of the 64 genes used in the analysis, 39 are associated with aggression in transgenic or knock-out studies of mice, and 25 genes are related to risky behavior such as drinking and drug use in the human population.

The results show that the proportion of variance in delinquency or violence explained by the 64 genes is considerably smaller among the married than among the unmarried, implying that marriage may suppress the collective genetic influence on antisocial behavior.