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.

GWAS have identified more than 700 genetic signals associated with type 2 diabetes (T2D). To gain insight into the underlying molecular mechanisms, we created the Translational human pancreatic Islet Genotype tissue-Expression Resource (TIGER), aggregating >500 human islet RNA-seq and genotyping datasets. We imputed genotypes using 4 reference panels and meta-analyzed cohorts to improve coverage of expression quantitative trait loci (eQTL) and developed a method to combine allele-specific expression across samples (cASE). We identified >1 million islet eQTLs (56% novel), of which 53 colocalize with T2D signals (60% novel). Among them, a low-frequency allele that reduces T2D risk by half increases CCND2 expression. We identified 8 novel cASE colocalizations, among which an SLC30A8 T2D associated variant. We make all the data available through the open-access TIGER portal ( http://tiger.bsc.es ), which represents a comprehensive human islet genomic data resource to elucidate how genetic variation affects islet function and translate this into therapeutic insight and precision medicine for T2D.

The use of large-scale supercomputing architectures is a hard requirement for scientific computing Big-Data applications. An example is genomics analytics, where millions of data transformations and tests per patient need to be done to find relevant clinical indicators. Therefore, to ensure open and broad access to high-performance technologies, governments, and academia are pushing toward the introduction of novel computing architectures in large-scale scientific environments. This is the case of RISC-V, an open-source and royalty-free instruction-set architecture. To evaluate such technologies, here we present the Variant-Interaction Analytics use case benchmarking suite and datasets. Through this use case, we search for possible genetic interactions using computational and statistical methods, providing a representative case for heavy ETL (Extract, Transform, Load) data processing. Current implementations are implemented in x86-based supercomputers (e.g. MareNostrum-IV at the Barcelona Supercomputing Center (BSC)), and future steps propose RISC-V as part of the next MareNostrum generations. Here we describe the Variant Interaction Use Case, highlighting the characteristics leveraging high-performance computing, indicating the caveats and challenges towards the next RISC-V developments and designs to come from a first comparison between x86 and RISC-V architectures on real Variant Interaction executions over real hardware implementations.

Type 2 diabetes (T2D) is a complex and clinically heterogeneous disease. Although clustering approaches have defined clinical subtypes, their genetic and molecular architectures remain poorly understood. Efforts using genome-wide association studies (GWAS) have been constrained by modest sample sizes and conservative significance thresholds, limiting subgroup resolution. Leveraging recent expansions in cohort scale and diversity, we developed a new analytical framework that integrates GWAS with Machine Learning and eXplainable Artificial Intelligence. Applied to the UK Biobank, this strategy enabled the first comprehensive genetic and molecular characterization of T2D subtypes, identifying 184 genes that define each subgroup’s molecular landscape and associated tissue profiles. In addition, the framework improved detection of T2D risk compared to current Polygenic Risk Scores. The findings underscore the importance of clinical stratification to uncover the complex pathophysiology of T2D and to pave the way for more precise prevention and treatment strategies.