Rupert Dodkins 1 , John R. Delaney 1 , Tess Overton 2 , Frank Scholle 2 , Alba Frias 3 , Elisa Crisci 3 ,Nafisa Huq 4 , Ingo Jordan 5 , Jason T. Kimata 6 , and Ilya G. Goldberg 1
1ViQi Inc., Santa Barbara, CA, 93117, United States
2Department of Biological Sciences North Carolina State University Raleigh, NC 27695, USA
3College of Veterinary Medicine, Department of Population Health and Pathobiology, North Carolina State University, Raleigh, NC 27695, United States
4Melbec Microbiology Ltd, Rossendale, Lancashire, BB4 4QJ, United Kingdom
5ProBioGen AG, Goethestr. 54, 13086 Berlin, Germany
6Department of Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX, United States
Abstract
Infectivity assays are essential for the development of viral vaccines, antiviral therapies and the manufacture of biologicals. Traditionally, these assays take 2–7 days and require several manual processing steps after infection. We describe an automated assay (AVIA TM ), using machine learning (ML) and high-throughput brightfield microscopy on 96-well plates that can quantify infection phenotypes within hours, before they are manually visible, and without sample preparation. ML models were trained on HIV, influenza A virus, coronavirus 229E, vaccinia viruses, poliovirus, and adenoviruses, which together span the four major categories of virus (DNA, RNA, enveloped, and non-enveloped). A sigmoidal function, fit to virus dilution curves, yielded an R 2 higher than 0.98 and a linear dynamic range comparable to or better than conventional plaque or TCID 50 assays. Because this technology is based on sensitizing AIs to specific phenotypes of infection, it may have potential as a rapid, broad-spectrum tool for virus identification.
Read the whole paper here.
Read the paper at BioRxiv here.