Kinship Decouple and Phenotype Selection (KDPS)

KDPS resolves cryptic relatedness in genetic studies through phenotype-aware subject selection that maximizes retention of valuable data for analysis.

Key Features:

• Phenotype-Aware Decoupling: KDPS separates related individuals by incorporating phenotype prioritization alongside relatedness scores, enabling retention of subjects with phenotypes of interest even when exclusions are necessary.
• Flexible Phenotype Accommodation: KDPS accommodates quantitative and categorical phenotypes, with customization options to either prioritize specific phenotypes or maximize overall sample size.
• Multiple Relatedness Metrics: KDPS processes both kinship and identity by descent scores for determining relationships between subjects.
• Biobank-Scale Computational Efficiency: KDPS processes biobank-scale studies within practical timeframes, demonstrating significant computational efficiency compared to previously published software.

Scientific Applications:

• Genetic Studies with Rare Phenotypes: KDPS retains valuable subjects in studies targeting uncommon and rare phenotypes where phenotype-naive relatedness breaking would cause critical data loss.
• Epidemiological Research: KDPS enhances precision in subject selection for epidemiological studies by preserving phenotypically important individuals within related groups.
• Biobank-Scale Cohort Analysis: KDPS enables efficient processing of large-scale genomic cohorts such as the UK Biobank for analyses requiring removal of cryptic relatedness.

Methodology:

KDPS uses a novel algorithm that considers relatedness scores (kinship or identity by descent) together with phenotype prioritization to select which related individuals to retain or exclude.