Usage of genomic data leads to discovery of biological, demographic, and healthcare insights, and is beginning to prove useful for translational applications. A hurdle to widespread application of certain findings from genomic data is their lack of transferability across cohorts, especially those that differ in ancestry from individuals that results were originally obtained in, as well as other personal characteristics, such as age and sex, or environmental exposures. This dissertation tackles these areas across multiple fronts, focusing on risk factors for cardiometabolic diseases. Cardiometabolic diseases, including cardiovascular diseases and type II diabetes, are responsible for the most deaths worldwide of any group of diseases, and their risk factors, including cholesterol levels, adiposity, blood pressure, and blood sugar levels, are all significantly heritable. Chapters 2 and 3 focus on a high-coverage whole-genome sequencing dataset of ~6,000 ethnically diverse African individuals (the A6K dataset), sampled from dozens of regions across sub-Saharan Africa, diverse in their environments, lifestyles, genetics, and demographic histories. In Chapter 2, I characterize genetic variation within the A6K dataset and compare it to public datasets. In Chapter 3, I characterize phenotypic variation for a broad range of cardiometabolic traits and their association with lifestyle characteristics, ethnic groupings, and genotypes in the A6K dataset to assess phenotypic diversity, discover novel associated genetic loci, and assess polygenic score (PGS) performance. Chapters 4 and 5 focus on work related to PGS performance for predicting body mass index (BMI), as adiposity is well-known for having differential heritability among cohorts even of similar ancestry and, consequently, varying PGS performance. In Chapter 4, using up to four cohorts (N>500,000) of European and African ancestry individuals with available genetic and linked phenotypic data of cardiometabolic traits, I assess how different risk factors affect BMI PGS, and how to harness this information to increase PGS performance. In Chapter 5, I vary parameters of PGS construction and quantify how these factors affect PGS performance. These results showcase the extensive genetic and phenotypic diversity related to cardiometabolic traits within Africa, the importance of incorporating data from ethnically diverse populations for conducting GWAS, and methods for improving polygenic score performance.