Thanks to a software inspired by the biological phenomenon of convergent evolution—the process by which different species independently develop similar traits to adapt to comparable environmental conditions—a team of researchers from the Department of Electronics, Information and Bioengineering – Politecnico di Milano, coordinated by Anna Bernasconi, has developed an innovative tool to predict future variants of SARS-CoV-2.
The project, carried out in collaboration with the Azienda Ospedaliero-Universitaria Pisana and the National Institute for Infectious Diseases “Lazzaro Spallanzani” IRCCS in Rome, led to the creation of ConvMut (short for “Convergent Mutations”), a software capable of identifying the virus’s evolutionary mechanisms and supporting the updating of vaccines and monoclonal antibodies.
ConvMut is freely accessible through the international GISAID Data Science Initiative platform, which hosts over 17 million SARS-CoV-2 genome sequences shared by thousands of laboratories worldwide. The software automatically explores convergent viral mutations, generating evolutionary charts that previously required manual analysis. It groups viral lineages based on sets of shared mutations—often acquired incrementally—and provides near real-time information on the evolution of SARS-CoV-2.
Although the COVID-19 pandemic is now largely under control thanks to vaccination campaigns, the virus continues to mutate, especially in immunocompromised patients, for whom it remains a serious threat. Each year, vaccines are updated according to the dominant strain, but the process from strain selection to vaccine distribution can take months—often too slow to keep pace with the virus’s evolution.
ConvMut aims to bridge this temporal gap, allowing researchers to predict months in advance which strain is likely to become dominant and to develop more targeted and effective vaccines.
Beyond vaccination, the software could also play a crucial role in the design of monoclonal antibodies, which are among the few viable therapies for vulnerable patients. Predicting how the Spike protein—the main target of both vaccines and monoclonals—will evolve can help pharmaceutical companies better focus their research and development efforts, reducing time, costs, and the risk of failure.