Shaila Musharoff
Assistant Professor
Make your next move.
Prepare your application
Review drafts before you send them.
Explore related searches
Articles (2)
Low-density lipoprotein cholesterol (LDL-C) is a well-established and modifiable risk factor for cardiovascular diseases (CVDs), which has been the leading cause of mortality among women and men in the United States since 1921. Despite research demonstrating sexually dimorphic symptom presentation of CVD, women remain underdiagnosed and undertreated for CVDs relative to men. Using genotype and phenotype data from the All of Us (AoU) Research Program, we examined sex-specific differences in LDL-C measurements, including baseline levels, statin treatment efficacy, age-related patterns, genome-wide association study (GWAS) results, and heritability between women and men. We find that the median LDL-C measurements of women are consistently higher than those of men across all age strata and that this difference is most apparent after 40 years of age. In addition, women are treated with statins at a lower rate than their male counterparts, and statins appear to be less effective in women—particularly Black women, who exhibited the highest median LDL-C levels while prescribed statins. Based on GWAS, genetic variants associated with LDL-C measurements differ between women and men, as do their effect sizes. Finally, heritability differs between sex, age, racial identity, and statin treatment groups. These findings indicate that current clinical intervals of LDL-C and pharmaceutical-based LDL-C modification approaches may not be equally appropriate across subgroups, with significant variation by sex and self-identified race. This highlights the need for sex-specific and population-informed strategies in both the treatment of LDL-C and genetic studies.
Year:
2026
Social factors of health (SFOH) are critical determinants of human traits but are rarely incorporated into genetic models. Here, we assess how participant-completed SFOH survey data influence heritability estimates and trait variance in 85,963 diverse All of Us participants spanning multiple ancestries and racial identities. We summarized SFOH survey data using multiple correspondence analysis (MCA) into axes representing distinct social dimensions (e.g., social support, perceived stress) and developed a Social Similarity Index (SSI) to capture social environment similarity among individuals. To analyze trait heritability, we used Haseman-Elston (HE) regression for its sensitivity to residual structure. Including SFOH measures as covariates in HE regression models significantly reduced heritability estimates for four of 18 traits, three of which are anthropometric: body mass index, hip circumference, waist circumference, and HDL cholesterol. This suggests that unmodeled social structure can be misattributed to genetic effects. Though several SFOH measures have non-zero heritability when adjusting for three genetic principal components ( h 2 ≈ 0.01 − 0.09), these are reduced to similar estimates when adjusting for seven genetic principal components ( h 2 ≈ 0.01 − 0.02). This convergence across SFOH measures, which capture different social dimensions, indicates that SFOH measures capture population stratification, emphasizing their utility for reducing confounding in heritability estimates. SFOH measures were also associated with trait variance, with anthropometric traits—those with reduced heritability—exhibiting the most extensive dispersion effects. This implies that social environment influences trait variability. Our findings highlight the necessity of including social and environmental factors in genetic studies to reduce potential confounding.
Year:
2025
Collaborators (2)
Nasa Sinnott-Armstrong
Assistant Professor
Fred Hutch Cancer Center
Oona Shigeno Risse-Adams
Cornell University
Make your next move.
Prepare your application
Review drafts before you send them.