Functional Data Analysis (FDA) is a fast-expanding body of methods and tools to analyze data comprised of curves or surfaces. Its applications in biomedical research are growing. This talk will feature two studies that illustrate its use. The first is a study of Childhood obesity leveraging a longitudinal phenotype measured on a small sample of individuals for identifying genetic variants that may predispose to fast weight gain in early childhood. The second is a study of data from the first wave of COVID-19 in Italy, characterizing a heterogeneous epidemic through the shapes of its curves across different regions, and relating mortality to mobility, positivity, and a number of demographic, socio-economic and infrastructural covariates.



The first study is joint work with a team of Penn State researchers, including S. Craig, A.M. Kenney, J. Lin, I.M. Paul, the late L.L. Birch, J. Savage, M.E. Marini, M.L. Reimherr and K.D. Makova. The second study is joint work with a team of researchers from both Penn State and the Sant’Anna School of Advanced Studies (Pisa, IT), including T. Boschi, J. Di Iorio, L. Testa and M.A. Cremona.