About the poster
At Bio-IT World 2024, Ardigen presented a novel approach to enhance Mode of Action (MoA) prediction by combining chemical structure and phenotypic screening data. Using deep learning techniques, the team developed a multimodal model that integrates features from High Content Screening (HCS) images and compound structures. This fusion significantly outperformed traditional, human-defined descriptors—demonstrating higher prediction accuracy and faster inference times.
The study leveraged open-source datasets and proprietary AI models to show that combining visual and structural data yields a synergistic effect, enabling more reliable drug discovery outcomes. This work marks a major step forward in phenotypic screening, showcasing Ardigen’s commitment to advancing AI-powered solutions in life sciences.
