Online Program

Return to main conference page

All Times EDT

Friday, September 24
Fri, Sep 24, 1:00 PM - 2:00 PM
Virtual
Poster Session II

The PADAG App: Power Analysis and Genetic Data Augmentation with Patients Missing Genotypic Data for Precision Medicine (302358)

Joshua Xu, US Food and Drug Administration 
*Wei Zhuang, US Food and Drug Administration 

Keywords: Precision Medicine, Missing Genotypes, Data Augmentation, Genotype Imputation, Family Studies, Web Application

Pharmacogenomics and genetic studies offer great promise in precision medicine and public health, improving scientific knowledge on how genes affect a person’s responses to certain drugs or exposures. In some real-world settings, probands revealed interesting adverse events or phenotypes, but passed away before providing DNA (due to aging, rapidly progressing lethal diseases, etc.) The phenotypic and genotypic data of their relatives are often available or accruable. For a simple and realistic occurrence with probands missing genotypes completely at random, we developed a software tool and made it publicly available for researchers to determine the power gain of including ungenotyped probands and genotyped relatives in the design and data analysis of their studies on continuous or dichotomous outcomes. The program asks users to select types of hypothesis tests and provide information based on prior knowledge regarding family-level and population-level parameters (e.g., heritability and allele of frequency). Empowered with a published power gain formula for a continuous outcome and a refined power gain formula for a dichotomous outcome, the PADAG (Power Analysis and Genetic Data Augmentation with Patients Missing Genotypic Data for Precision Medicine) App provides information on augmented power if phenotyped but ungenotyped patients are included in study design and analysis, in addition to phenotyped and genotyped relatives. User-friendly interfaces are provided for researchers to easily input baseline information and obtain and visualize augmented statistical power for tests on alleles of different genes. Our results show that the inclusion of ungenotyped probands in study design and data analysis can help enhance the discovery of real-world evidence on the effects of genetic variants on biological outcomes or responses, such as to toxicity and infectious agents, for precision medicine and healthcare.