The identification and characterisation of genomic changes (variants) that can lead to human diseases is one of the central aims of biomedical research. The generation of catalogues of genetic variants that have an impact on specific diseases is the basis of Personalised Medicine, where diagnoses and treatment protocols are selected according to each patient’s profile. In this context, the study of complex diseases, such as Type 2 diabetes or cardiovascular alterations, is fundamental. However, these diseases result from the combination of multiple genetic and environmental factors, which makes the discovery of causal variants particularly challenging at a statistical and computational level. Genome-Wide Association Studies (GWAS), which are based on the statistical analysis of genetic variant frequencies across non-diseased and diseased individuals, have been successful in finding genetic variants that are associated to specific diseases or phenotypic traits. But GWAS methodology is limited when considering important genetic aspects of the disease and has not yet resulted in meaningful translation to clinical practice. This review presents an outlook on the study of the link between genetics and complex phenotypes. We first present an overview of the past and current statistical methods used in the field. Next, we discuss current practices and their main limitations. Finally, we describe the open challenges that remain and that might benefit greatly from further mathematical developments.