RESUMO
Morphological profiling of cells in the presence of perturbants, also known as phenomics, is gaining momentum given its successful implementation for drug discovery and compound profiling. The current COVID-19 pandemic has fueled the search for new and fast methods to identify novel or repurposed therapeutic drugs. A popular method to identify antiviral drugs is the use of antibody-based immunofluorescence to visualise infected cells. However, this method lacks depth towards the effect of such drugs on the host cells. Here we present a phenomics workflow for untargeted phenotypic drug screening of virus infected cells, combining Cell Painting with antibody-based detection of viral infection in a single and simple method and provide a semi-automated image analysis pipeline for classification and feature extraction of virus infected cells. Our phenomics workflow provides valuable information about the effect of both virus and drugs on the host cells. We validated our method using a panel of 9 antiviral compounds including known and novel compounds on MRC5 human lung fibroblasts infected with Human coronavirus 229E (CoV-229E). Two of the compounds showed strong antiviral efficacy concomitant with a recovery of the morphological profile towards non-infected.