A research coordinated by Siena has been published on Human Genetics
With 40 Italian hospitals created a mathematical model of machine learning to predict the severity of Covid-19
SIENA. An all-Italian work has been published in the prestigious international journal Human Genetics, coordinated by Siena together with 40 other Italian hospitals, which develops a mathematical learning model to predict the severity of Covid based on genetic analysis. The study was coordinated by Professor Alessandra Renieri, full professor of Genetics at the University of Siena and director of Medical Genetics of the Sienese university hospital and is the result of research over the last 20 months and the collaboration of professionals from Aou Senese who yes others from Covid, together with other colleagues 40 Italian hospitals.
The work demonstrates how from the analysis of the genes, carried out on 4 thousand positive samples, it was possible to define a mathematical model of machine learning that can be interpreted to predict the severity of Covid-19. The pattern was confirmed internationally of European ethnicity. The researchers have developed an experimental genetic test – used for research purposes and not for assistance – which uses the most advanced sequencing technologies, through the NovaSeq 6000 instrument, combined with special analysis techniques, designed by “data scientists” and artificial intelligence experts including Simone Furini and Marco Gori and doctoral student Nicola Picchiotti. The importance of collaboration between different researchers was underlined by Dr. Chiara Fallerini, co-first name of the work and researcher in Medical Genetics: “It was a huge undertaking – he said – based on tenacity and collaboration between very distant disciplines” .
“Finally we have a mathematical model that represents the evolution of the inheritance rules postulated by Mendel 155 years ago – comments Professor Alessandra Renieri who is coordinator of the Italian Gen-Covid consortium -. Genetic evaluation is based on DNA sequencing with the NovaSeq platform and has advantages: calculation of probabilities in non-infected subjects through a numerical severity score on the genetic makeup of the many genes involved in the pathogenesis of Covid; in the already infected and severe subject it triggers can indicate which are the weak points such as genes cheno thrombosis, or the capacity of the anti-inflammatory effect of testosterone in males, or even lack of stimulus from the receptors that should “feel” the virus with consequence lack of innate immune response. We are a professor with AIFA – Renieri concluded the week – for the approval of a trial platform for personalized adjuvant therapy in severe Covid patients based on genetic markers and, in the coming months, we hope to be able to improve the treatment “.