“During my stay at BT, I had the opportunity to get a deeper understanding of high throughput NMR metabolomics thanks to the expertise of BT colleagues and to a visit to the wet lab facility Center of Omics Science in Reus, where I was able to see the equipment and part of the upstream data processing. Further, together with the BT team I started investigating a clinical metabolomics data set to get a grasp on this specific kind of data, which will be available for the CardioSCOPE cohort later in the project. This metabolomics data set acted thus as a test bed for the initial implementation of a data science workflow, which includes an unsupervised learning step to discover patient subgroups, and a supervised learning step to find a shortlist of predictors associated with a clinical outcome. Part of this preliminary analysis was dedicated to defining Standard Operating Procedures for machine learning.”
Marco Chierici, Senior Data Scientist & Bioinformatician at the Bruno Kessler Foundation