MAVISp: Multi-layered Assessment of VarIants by Structure for proteins

The role of genomic variants in disease, including cancer, continues to be realized. Thanks to advances in sequencing techniques and their introduction in clinical practice, the number of genomic variants discovered is growing at fast pace. Many variants are classified as Variants of Uncertain Significance (VUS) or conflicting evidence, posing big challenges to their interpretation and usability. We here introduce a modular structure-based framework for the characterization of the impact of protein variants (i.e., MAVISp, M ulti-layered A ssessment of V ar I ants by S tructure for p roteins), accompanied by a Streamlit-based web server ( https://services.healthtech.dtu.dk/services/MAVISp-1.0/ ) to facilitate data access, consultation, and reusability. Currently, MAVISp includes analyses of 167 different proteins and approximately 50 000 variants. New protein targets are routinely analyzed in batches by a team of biocurators through standardized workflows, using high-throughput free energy calculations and simulations. We also illustrated the potential of the approach on a selection of case studies. New variants will be deposited regularly or in connection with future publications where the framework will be applied.