Relation Therapeutics and GSK announced an expanded research collaboration to generate large-scale cellular perturbation datasets for cellular foundation models. Under the terms, Relation could receive up to $110 million in upfront and milestone payments, with the work centered on producing petascale multi-omic perturbation data. The deal follows Relation’s launch of MORGAN (Multi-Omic Regulatory Genomics using Artificial Neural Networks), described as a general-purpose model for predicting how cells respond to genetic and pharmacological interventions across disease contexts. Each tissue-specific MORGAN instance will be built to learn intervention responses relevant to specific cell types implicated in disease. For GSK, the collaboration is another step toward using automated laboratory generation of high-consistency perturbation data to accelerate target discovery, while Relation’s model-building approach keeps the output oriented toward therapeutic target selection rather than purely descriptive biology.