Deep learning models and CRISPR-based gene knockout techniques uncovered 998 host genes involved in Ebola virus replication. To identify potential therapeutic targets, researchers combined artificial intelligence (AI), image-based genome-wide CRISPR screening, and single-cell imaging technologies. AI-powered autoencoders and machine learning algorithms were used to classify different stages of viral infection and identify critical host factors, including UQCRB and STRAP. Notably, inhibition of UQCRB using a small-molecule compound significantly reduced Ebola infection in vitro, highlighting the promising potential of AI-driven precision antiviral drug discovery for emerging infectious diseases.