Sex-differential allele frequencies at candidate cardiometabolic loci in a multi-ancestry reference cohort: an exploratory analysis with external comparison

Background Polygenic risk scores for cardiovascular disease show poor agreement between methods, and most treat sex as a covariate rather than stratifying by it, despite well-documented sex differences in disease incidence and outcomes. We asked whether sex-differential allele frequencies at candidate cardiometabolic loci are detectable, and tested the robustness of any signal using ancestry adjustment and external data. Methods Sex-stratified allele frequencies were compared across five candidate regions (APOE, 9p21, CRP, ADIPOQ/CDH13, IL-6) in 2,483 individuals from five ancestral groups in the 1000 Genomes Project Phase 3. Variants differing by more than 2% between sexes were identified, and five lead variants were combined into a composite score. Three further analyses tested robustness: regression adjusted for ten genetic principal components, within-ancestry stratified testing, and comparison with gnomAD v4.1.1. The chromosome 19 extraction was run chromosome-wide rather than restricted to the APOE window, so the APOE region was also analysed separately. Results A total of 182 sex-differential variants were identified across four regions, most extensively at CDH13 (67 variants). The composite score was 17.2% higher in female participants ( p = 4.7 × 10⁻⁸, Cohen’s d = 0.22), and directionally consistent across all five ancestral groups. Because the lead variants were selected for this property in this cohort, however, that difference is expected by construction and is not independent confirmation. Robustness analyses were mixed. After ancestry adjustment and Bonferroni correction for five tests, two variants retained support (rs7258841, p = 9.4 × 10⁻⁵; rs111339851, p = 0.0025). Within-ancestry testing gave a consistent direction of effect but only 2 of 25 tests reached nominal significance. In gnomAD, 4 of 25 tests were nominally significant: rs7258841 was concordant in two ancestral groups, while rs111339851 showed the opposite direction in four of five. Separate analysis of the APOE region found no variant exceeding the 2% threshold and none significant after correction, including the two variants defining the ε2/ε3/ε4 alleles. Conclusions Sex-differential allele frequencies are detectable in this cohort but only partially survive ancestry adjustment and are inconsistently supported externally. These findings are hypothesis-generating, do not support clinical risk estimation, and require validation in outcome-linked cohorts.

Authors

Publication Details

Journal
Biology of Sex Differences
Published
2026-11-19
DOI
https://doi.org/10.1186/s13293-026-01019-z
Primary Topic
Genetic Associations and Epidemiology
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Sex-differential allele frequencies at candidate cardiometabolic loci in a multi-ancestry reference cohort: an exploratory analysis with external comparison

Vanathi Gopalakrishnan, Smmrithi Ravindran
Biology of Sex Differences
Genetic Associations and Epidemiology
article

Sex-differential allele frequencies at candidate cardiometabolic loci in a multi-ancestry reference cohort: an exploratory analysis with external comparison

Vanathi Gopalakrishnan, Smmrithi Ravindran
article en

Abstract

Background Polygenic risk scores for cardiovascular disease show poor agreement between methods, and most treat sex as a covariate rather than stratifying by it, despite well-documented sex differences in disease incidence and outcomes. We asked whether sex-differential allele frequencies at candidate cardiometabolic loci are detectable, and tested the robustness of any signal using ancestry adjustment and external data. Methods Sex-stratified allele frequencies were compared across five candidate regions (APOE, 9p21, CRP, ADIPOQ/CDH13, IL-6) in 2,483 individuals from five ancestral groups in the 1000 Genomes Project Phase 3. Variants differing by more than 2% between sexes were identified, and five lead variants were combined into a composite score. Three further analyses tested robustness: regression adjusted for ten genetic principal components, within-ancestry stratified testing, and comparison with gnomAD v4.1.1. The chromosome 19 extraction was run chromosome-wide rather than restricted to the APOE window, so the APOE region was also analysed separately. Results A total of 182 sex-differential variants were identified across four regions, most extensively at CDH13 (67 variants). The composite score was 17.2% higher in female participants ( p = 4.7 × 10⁻⁸, Cohen’s d = 0.22), and directionally consistent across all five ancestral groups. Because the lead variants were selected for this property in this cohort, however, that difference is expected by construction and is not independent confirmation. Robustness analyses were mixed. After ancestry adjustment and Bonferroni correction for five tests, two variants retained support (rs7258841, p = 9.4 × 10⁻⁵; rs111339851, p = 0.0025). Within-ancestry testing gave a consistent direction of effect but only 2 of 25 tests reached nominal significance. In gnomAD, 4 of 25 tests were nominally significant: rs7258841 was concordant in two ancestral groups, while rs111339851 showed the opposite direction in four of five. Separate analysis of the APOE region found no variant exceeding the 2% threshold and none significant after correction, including the two variants defining the ε2/ε3/ε4 alleles. Conclusions Sex-differential allele frequencies are detectable in this cohort but only partially survive ancestry adjustment and are inconsistently supported externally. These findings are hypothesis-generating, do not support clinical risk estimation, and require validation in outcome-linked cohorts.

Biology of Sex Differences
Genetic Associations and Epidemiology
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon