QGS

Quantitative Genetic Scoring (QGS): Dimensionality reduction for genomic data

Quantitative Genetic Scoring (QGS) generates quantitative genetic variables for arbitrary genomic regions by computing the sum of absolute differences between an individual’s genetic sequence and a reference population, producing a genetic distance metric suitable for integration into statistical models.

Key Features:

• Dimensionality Reduction: Condenses genome-wide data into region-level quantitative scores while retaining phenotypically relevant genetic variance.
• Genetic Distance Metric: Calculates the sum of absolute sequence differences relative to a reference population.
• Phenotypic Variance Retention: Reduces genetic information by >98% while preserving phenotypic variance at low, medium, and high granularity levels.
• Independence from Region Size and Linkage Disequilibrium: Produces associations independent of genomic region size and linkage disequilibrium structure.
• Compatibility with Stability Selection: Enhances detection of significant associations when combined with stability selection compared to conventional genome-wide association studies (GWAS).

Scientific Applications:

• Integrated Genomic Modeling: Incorporates quantitative genetic region scores into statistical analyses of complex traits and diseases.

Methodology:

QGS computes region-specific scores by summing absolute genotype differences between individuals and a reference population to derive genetic distance measures, enabling large-scale dimensionality reduction while maintaining phenotypic signal and statistical robustness independent of linkage disequilibrium and region size.