Related Experiment Video
Updated: Aug 7, 2026
10:16
High-throughput Antiviral Assays to Screen for Inhibitors of Zika Virus Replication
Published on: October 30, 2021
AI-driven Identification of Zika Virus Inhibitors
Yuka Otsuka1, Justin Shumate1, Teresa Findley Rye2
1Herbert Wertheim UF Scripps Institute for Biomedical Innovation & Technology, Department of Molecular Medicine, Jupiter, FL 33458.
SLAS Discovery : Advancing Life Sciences R & D
|August 5, 2026
Summary
This study developed an AI-driven high throughput screening assay to identify Zika virus inhibitors. The AI method rapidly detected viral effects, improving cost- and time-effectiveness in drug discovery.
Area of Science:
- Computational Biology and Drug Discovery
- Virology and Infectious Diseases
- Artificial Intelligence in Medicine
Background:
- Zika virus (ZIKV) poses significant health risks, particularly microcephaly in infants, with no current vaccines or antiviral treatments.
- High Throughput Screening (HTS) is crucial for drug discovery but can be time-consuming and costly.
- Advancements in Artificial Intelligence (AI) and imaging technologies offer potential for more efficient HTS.
Purpose of the Study:
- To develop and validate an AI-driven HTS assay for identifying ZIKV inhibitors.
- To combine a miniaturized cytopathic effect (CPE) assay with advanced AI image analysis for rapid ZIKV drug screening.
- To assess the cost- and time-effectiveness of AI-powered HTS in antiviral drug discovery.
Main Methods:
- Developed a miniaturized 1536-well plate cell viability-based CPE assay for ZIKV.
- Screened 1,280 unique compounds using the assay and an ultra-high throughput imager.
- Analyzed high-throughput images using two AI platforms: AVIA (deep convolutional neural networks) for infectivity scoring and AutoHCS for phenotypic profiling.
Main Results:
- AVIA accurately distinguished infected from uninfected cells as early as 40 hours post-infection, significantly reducing assay duration.
- AutoHCS phenotypic profiling combined with AVIA scores identified compounds with potential antiviral effects.
- 12 out of 20 hit compounds from the original assay were confirmed by AVIA and AutoHCS, demonstrating method reliability and concordance.
- AI analysis allowed for the elimination of false-positive compounds without additional assays.
Conclusions:
- The developed AI-driven HTS assay is robust, sensitive, and cost-effective for ZIKV inhibitor identification.
- Combining AI tools like AVIA and AutoHCS with high-quality imaging data streamlines the drug discovery process.
- This integrated approach significantly enhances the efficiency and reliability of HTS in identifying potential antiviral therapeutics.
