Covariate selection for association screening in multiphenotype genetic studies.

Area of Science:

• Genetics
• Statistical genetics
• Bioinformatics

Background:

• Big data analysis, particularly in human genetics, struggles with multiple comparisons, obscuring true associations.
• Current methods for detecting weak genetic associations rely on marginal statistical approaches and large sample sizes.
• Existing strategies do not fully utilize correlated environmental and genetic factors present across multiple phenotypes in cohort studies.

Purpose of the Study:

• To develop a novel statistical approach for enhancing the power of genetic association testing in studies with multiple correlated phenotypes.
• To introduce covariates for multiphenotype studies (CMS) as a method to improve signal detection.
• To demonstrate the efficacy of CMS in real and simulated datasets.

Main Methods:

• Development of the covariates for multiphenotype studies (CMS) framework.
• Application of CMS to analyze correlated phenotypes within large-scale human genetic datasets.
• Validation using both simulated data and real-world cohort data.

Main Results:

• CMS effectively leverages correlated phenotypes to increase statistical power in association testing.
• The power gains achieved with CMS can exceed those from a twofold increase in sample size.
• Analyses confirmed the utility of CMS in both simulated and real genetic association studies.

Conclusions:

• Correlated phenotypes offer a valuable, underutilized resource for boosting power in genetic association studies.
• The CMS approach provides a statistically sound method to harness this resource.
• Implementing CMS can lead to more efficient and powerful discovery of genetic associations, especially for variants with small effects.