#IDIBELLseminars: Cancer Genome Interpreter: Identifying the drivers of cancer to improve genomic interpretation

The interpretation of cancer genomic data is crucial for Precision Oncology. However, the high prevalence of Variants of Unknown Significance (VUS) poses major challenges. To address this, we developed an Automatic Learning Platform to identify cancer driver genes and mutations by learning from thousands of tumor genomes.
This computational framework includes Intogen, which identifies 633 driver genes across 87 cancer types, and BoostDM, a machine learning method that classifies all possible SNVs as driver or passenger. BoostDM contains over 500 gene- and tumor-type-specific predictive models enabling the systematic annotation of VUS. This platform is integrated into the Cancer Genome Interpreter (CGI), a tool that combines these predictions with expert-curated databases to interpret mutations from cancer patients, identifying oncogenic mutations and biomarkers of drug response. CGI is being enhanced in clinical settings to better support data-driven and personalized treatment decisions in precision oncology.

Hosted by Laura Costas – Infections and Cancer group

Education: Bachelor’s degree in Biology (University of Barcelona), Master’s in Biomedical Research, Master’s in Bioinformatics and Biostatistics, PhD in Biomedicine.
Previous Experience: PhD at the Centre for Genomic Regulation (CRG, Barcelona), Postdoctoral researcher at Hospital Clínico de Barcelona.
Other: Collaborating lecturer at the University of Barcelona and other universities.
Current Position: Senior Researcher at Barcelona Biomedical Genomic Lab in the Institute for Biomedical  Research (IRB Barcelona) and Scientific Coordinator of the European project Cancer Genome Interpreter Clinics (CGI-Clinics).