Polygenic Risk Scores (PRS) have been used extensively in recent years to estimate an individual’s genetic risk of developing a complex chronic disease. For their calculation, results from genomics studies are used and both the effect of each studied polymorphism and its frequency in the population are taken into account. Their role as a prognostic indicator for the development of a certain disease and their use in clinical practice requires further extensive studies, while a limiting factor is that most data from broad genomic studies come only from populations of European origin. (Kunkle, et al., 2019; Choi, et al., 2019; Mavaddat, et al., 2019; Ge, et al., 2019; Lello, et al., 2018; Evans, et al., 2018).
The purpose of this PhD thesis is to estimate PRS for various complex diseases and conditions, in large databases, in order to assess their prognostic ability as well as their applicability to populations such as Africans or Asians. For their calculation, databases such as UK-Biobank will be used which has ~500,000 fully genotyped individuals with additional available rich phenotypic and other clinical characteristics. UK-Biobank participants are mainly of European origin, but a few thousand people of African or Asian descent also participate.
For our analysis we will use various proposed statistical methods to produce polygenic risk estimates (PRS) such as the following: PC+T method, SBLUP method, Ldpred2-Inf method, LDpred-funct method, LDpred2 method, Lassosum method, PRS- method CS, PRS-CS-auto method, SBayesR method and MegaPRS method among others (Mullins, et al., 2019; Khera, et al., 2018; Du Rietz, et al., 2018; Dudbridge, et al., 2018 ; Torkamani, et al., 2018)
