Integrating Multi-Modal High-Content Screening Data with Machine Learning for Drug Discovery and Target Identification

High-content screening (HCS) is at the forefront of drug discovery, offering multi-modal datasets that unite rich imaging, diverse perturbations, and deep molecular profiling. As the field transitions toward more comprehensive assays—encompassing CRISPR/RNAi, small molecules, genomics, proteomics, and contextual measurements—there is unprecedented potential for revealing subtle cellular phenotypes, mechanisms of action, and druggable targets. However, this explosion of complex data comes with notable challenges: batch effects, technical variation, missing modalities, and the inherent difficulty of linking intricate phenotypic data to interpretable, actionable biological mechanisms.This Research Topic aims to drive innovation in computational, statistical, and machine learning solutions that transform HCS data into actionable insights for drug discovery and target identification. The focus is on enabling robust, interpretable, and generalizable methods, including approaches that connect phenotypic signals to biological pathways, networks, and structure-based information. We welcome contributions that demonstrate potential both for early-stage hit identification and for advancing mechanistic understanding, with a clear link to practical prioritization and validation in experimental pipelines.To gather further insights into integrative, multimodal high-content screening and machine learning for drug discovery, we encourage submissions that focus on a defined range of topics while emphasizing reproducibility and translational value. We welcome articles addressing, but not limited to, the following:• Multimodal modeling and handling of missing data types• Representation learning for phenotype profiling and discovery• Robustness and domain adaptation across experimental settings• Mechanism- and target-inference using networks, knowledge graphs, and graph learning• Analysis of dose–response, time-series, and dynamic screens• Causal inference and confounding control• Uncertainty-aware prioritization and decision support• Interpretability and experimental validation strategies• Benchmarks, datasets, and reproducible computational pipelines• Integration with structure-based modeling or drug repurposing, when directly informed by HCS resultsWe invite Original Research, Methods, Data Report, Technology and Code, Review, and Mini Review articles. Topic editors: Zuyi (Jacky) Huang, Mahendiran Dharmasivam, Busra Kaya Source: https://www.frontiersin.org/research-topics/78383/integrating-multi-modal-high-content-screening-data-with-machine-learning-for-drug-discovery-and-target-identification