Year: 2025
Type: article
Abstract: Image-based profiling is rapidly transforming drug discovery, offering unprecedented insights into cellular responses. However, experimental variability hinders accurate identification of mechanisms of action (MoA) and compound targets. Existing methods commonly fail to generalize to novel compounds, limiting their utility in exploring uncharted chemical space. To address this, we present a confounder-aware foundation model integrating a causal mechanism within a latent diffusion model, enabling the generation of balanced synthetic datasets for robust biological effect estimation. Trained on over 13 million Cell Painting images and 107 thousand compounds, our model learns robust cellular phenotype representations, mitigating confounder impact. We achieve state-of-the-art MoA and target prediction for both seen (0.66 and 0.65 ROC-AUC) and unseen compounds (0.65 and 0.73 ROC-AUC), significantly surpassing real and batch-corrected data. This innovative framework advances drug discovery by delivering robust biological effect estimations for novel compounds, potentially accelerating hit expansion. Our model establishes a scalable and adaptable foundation for cell imaging, holding the potential to become a cornerstone in data-driven drug discovery.
Source: npj Imaging
Language: English
Subfield: Biophysics
Domain: Life Sciences
Open Access status: gold
Estimated APC paid: $2,790
Funders: National Science Foundation, Pfizer, Nvidia, Wellcome Trust, UK Research and Innovation, National Institute for Health and Care Research, Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung, National Natural Science Foundation of China, Rural Development Administration, Horizon 2020 Framework Programme, HORIZON EUROPE Framework Programme, Medical Research Council, NIHR Imperial Biomedical Research Centre
Awards: G1251990094, G1972488856, G2597775472, Innovative AI-Empowered Organoid Platform for Illuminating Early Neural Tube Development and Related Neural Tube Defects, Fully Automatic Segmentation and Assessment of Atrial Scars for Atrial Fibrillation Patients
Using LGE MRI, lnvestigating a multidrug-resistant pathogen through the lens of T cell immunity, G6756521213, G7986630670, Synergistic Inverse Problems Omni-Solver for Expeditious High Quality Multimodal Cardiovascular MRI via Deep Compressive Sensing and Data Coalescing, Novel Data-Driven Models for robust and generalisable ultra-low-dose PET image reconstruction and high-quality CT-free PET image synthesis
OpenAlex ID: W4415418703