Multi-polygenic score approach to trait prediction.

Area of Science:

• Genetics
• Developmental Biology
• Biostatistics

Background:

• Polygenic scores aggregate DNA variants from genome-wide association studies (GWASs) to estimate genetic predispositions.
• Traditional polygenic scores often use a single GWAS, limiting predictive capacity for complex traits.

Purpose of the Study:

• To introduce and evaluate a multi-polygenic score (MPS) approach for enhanced prediction of developmental outcomes.
• To assess the predictive power of MPS compared to single polygenic scores for educational achievement, body mass index (BMI), and general cognitive ability.

Main Methods:

• Utilized summary statistics from 81 GWASs across cognitive, medical, and anthropometric traits.
• Employed regularized regression with cross-validation to select and combine 81 polygenic scores.
• Validated the MPS approach in an independent UK sample of 6710 adolescents.

Main Results:

• The MPS approach explained 10.9% of variance in educational achievement, 4.8% in general cognitive ability, and 5.4% in BMI.
• MPS improved prediction by 1.1% for educational achievement, 1.1% for general cognitive ability, and 1.6% for BMI compared to the best single-score models.
• Demonstrated increased predictive accuracy by integrating multiple GWAS datasets.

Conclusions:

• The multi-polygenic score (MPS) approach offers superior predictive power for developmental outcomes compared to single polygenic scores.
• MPS is adaptable and can incorporate future GWAS findings to maximize phenotype prediction.
• This method holds potential for research on developmental issues, gene-environment interactions, and future clinical applications in personalized medicine.