Integrated genomic and BMI analysis for type 2 diabetes risk assessment

Area of Science:

• Genetics and Genomics
• Epidemiology
• Biostatistics

Background:

• Type 2 Diabetes (T2D) is a growing global health concern, linked to genetic and environmental factors, including the obesity epidemic.
• Genome-Wide Association Studies (GWAS) have identified numerous genetic variants associated with T2D risk.
• Integrating genetic information into risk assessment models may enhance early identification of at-risk individuals.

Purpose of the Study:

• To evaluate the impact of incorporating genetic data (SNPs) and Body Mass Index (BMI) into risk assessment models for Type 2 Diabetes (T2D).
• To compare the predictive performance of Logistic Regression and Neural Network models with and without genetic and BMI information.
• To determine the extent to which genetic variants and BMI improve T2D prediction accuracy.

Main Methods:

• Utilized phenotypic and genetic data from 5245 subjects (4306 controls, 939 cases) from the Framingham Heart Study Original and Offspring cohorts.
• Developed risk assessment models using Logistic Regression and Bayesian Regularized Neural Networks.
• Included covariates such as gender, exposure time, cohort, BMI, and 65 T2D-associated Single Nucleotide Polymorphisms (SNPs), assessed via ten-fold cross-validation.

Main Results:

• Inclusion of genetic information increased predictive ability by 2% compared to baseline models.
• Models incorporating BMI showed a 6% improvement in Area Under the Curve (AUC) from Receiver Operating Characteristic (ROC) analysis.
• The highest AUC (0.75) was achieved by a model combining BMI and a genetic score derived from 65 T2D-associated SNPs.

Conclusions:

• Genetic information and BMI significantly enhance the predictive accuracy of Type 2 Diabetes risk assessment models.
• Both Logistic Regression and Neural Network models benefit from the inclusion of genetic variants and BMI.
• These findings support the use of combined genetic and clinical data for improved T2D risk stratification and personalized prevention strategies.