# SNPs Filtered by Allele Frequency Improve the Prediction of Hypertension Subtypes

**Authors:** Yiming Li, Sanjiv J. Shah, Donna Arnett, Ryan Irvin, Yuan Luo  
**arXiv:** [2111.10471](https://arxiv.org/abs/2111.10471)  
**Date:** 2021-11-19  
**Summary page:** https://www.summarizepaper.com/en/paper/2111.10471v1/

## Key points

- Hypertension is a significant global health concern and the leading cause of cardiovascular disease and premature death.
- Personalized approaches are necessary for managing hypertension due to variations in prognosis and treatment among subtypes.
- Risk factors for hypertension include genetic predisposition and environmental influences.
- Researchers developed classification models for hypertension subtypes using environmental variables and selected genetic features.
- The use of allele frequency-filtered SNPs improved prediction accuracy in the models.
- SHapley Additive exPlanations (SHAP) values were used to identify key predictors in subtype classification.
- Age, race, sex, smoking status, and specific SNPs were identified as crucial predictors in an effective model.
- Older age and African American ethnicity were associated with increased risk across all three classes of hypertension.

## Summary

<Hypertension, Global Health Concern, Personalized Approaches, Genetic Landscape, Environmental Influences>

Hypertension is a significant global health concern and the leading cause of cardiovascular disease and premature death. To effectively manage this condition, personalized approaches are necessary due to variations in prognosis and treatment among hypertension subtypes. Risk factors for hypertension include genetic predisposition and environmental influences, which interact to contribute to its development. In a study on African American and European American populations, researchers developed classification models for hypertension subtypes by incorporating both environmental variables and selected genetic features. The use of allele frequency-filtered SNPs improved prediction accuracy in these models. SHapley Additive exPlanations (SHAP) values were used to identify key predictors in subtype classification. The most effective model included age, race, sex, smoking status, and specific SNPs as predictors. Age and race were consistently identified as crucial predictors across all three classes of hypertension, with older age and African American ethnicity associated with increased risk. This research contributes valuable insights into the complex interplay between genetic factors and environmental influences in determining hypertension subtypes. It highlights the importance of considering both genetic and environmental factors in understanding the pathogenesis of hypertension for improved patient outcomes.