Reading genomic variants opens the way to predictive medicine

Research by Genevan scientists on gene expression and the non-coding genome is a significant breakthrough for the future of personalized medicine.
Geneticists have taken an important step towards true predictive medicine by exploring the links between disease and genetic activity in different tissues. They thus constructed a model, the first step in identifying sequences in the non-coding genome indicating a disease-related pathogenic effect. In a second study, they went even further by associating the risk of developing a disease – in particular schizophrenia, cardiovascular diseases or diabetes – with the variability of the activity of the genome in different types of cells. And their results brought some surprises. Their findings may well revolutionize how each of us, according to his genome, will take care of his health in the future.
These studies are based on data from the international GTEx project, for "Genotype-Tissue Expression", launched in 2010 and co-directed by Professor Emmanouil Dermitzakis, geneticist at the Faculty of Medicine of the University of Geneva (UNIGE) and director of the Health 2030 Genomics Center. The objective of this project was to collect as many tissues as possible from a large number of individuals to understand the effects of genes and their variations. The data published over the last 7 years have allowed scientists worldwide to make considerable progress in analysing genomic variations specific to each of these tissues and predispositions to diseases.
Examining different types of human tissue from hundreds of people has led to a better understanding of how genomic variants – those changes in the spelling of DNA code inherited from our parents – could control how, when, and how many genes are activated and deactivated in different tissues, increasing the risk of developing a wide range of diseases. One of the main discoveries of the GTEx consortium is that the same variant present in multiple tissues may have a different effect depending on the tissue involved. A variant that affects the activity of two genes associated with blood pressure will, for example, have a greater impact on the expression of these genes in the tibial artery, even if the activity of the genes is higher in other tissues. .
To evaluate the influence of variants on gene activity, the researchers perform an analysis called "eQTL". An eQTL – or quantitative locus of expression of the characters – consists of an association between a variant at a specific location of the genome and the level of activity of a gene in a particular tissue. By comparing the eQTLs of different tissues to the genes associated with diseases one can therefore determine which tissues are most related to a disease. But if we can associate a region of the genome with a phenotype (a disease, for example), scientists were not yet able to determine exactly which nucleotide – the bricks of our DNA – when it mutates, contributes to the phenotype. question. Emmanouil Dermitzakis emphasizes as follows: "We needed to design a model to precisely link variants to a particular disease. Our goal, to simplify, was to locate the exact nucleotide that, in case of mutation, increases the risk of a disease, rather than the associated region or gene.
To build a solid model, scientists performed eQTL analyzes of hundreds of samples and identified thousands of causal variations in the non-coding genome. Using this dataset, they began building models to recognize these variations from DNA sequences, without linking them to existing phenotypes. As described by Andrew A. Brown, assistant professor in the Department of Genetic Medicine and Development of UNIGE’s Faculty of Medicine and one of the first authors of these studies: "We wanted to recognize pathogenic variants without any other information than this. sequence. If our model is confirmed, we will solve one of the major problems of modern genomics: by simply reading non-coding DNA sequences, we will be able to identify their pathogenic effects. This is the real future of predictive medicine.

University of Geneva
www.unige.ch/medecine/fr/carrousel/la-lecture-des-variants-genomique-ouvre-la-voie-a-la-medecine-predictive/