Rapid Deployment\nof Antiviral Drugs Using Single-Virus\nTracking and Machine Learning

The outbreak of emerging acute viral diseases urgently\nrequires\nthe acceleration of specialized antiviral drug development, thus widely\nadopting phenotypic screening as a strategy for drug repurposing in\nantiviral research. However, traditional phenotypic screening methods\ntypically require several days of experimental cycles and lack visual\nconfirmation of a drug’s ability to inhibit viral infection.\nHere, we report a robust method that utilizes quantum-dot-based single-virus\ntracking and machine learning to generate unique single-virus infection\nfingerprint data from viral trajectories and detect the dynamic changes\nin viral movement following drug administration. Our findings demonstrated\nthat this approach can successfully identify viral infection patterns\nat various infection phases and predict antiviral drug efficacy through\nmachine learning within 90 min. This method provides valuable support\nfor assessing the efficacy of antiviral drugs and offers promising\napplications for responding to future outbreaks of emerging viruses.