Relation Therapeutics

Relation’s science is organized around (1) multi-modal patient data, (2) perturbational omics and translational cellular models, (3) high-performance experimentation and computation (including a partnership with NVIDIA for hyperscale compute), and (4) machine learning methods spanning target identification, prioritisation, validation, and experimental design. In terms of products/platforms, Relation positions multiple named internal systems as key differentiators: an integrated “wet–dry laboratory” (to couple human single-cell multi-omics and computational prediction with automatic experiment design), “Selectomatic,” an indication-selection machine-learning engine referenced in company announcements around the selection of osteoporosis as its first indication, and “Osteomics,” described as its proprietary functional single-cell bone atlas derived from patient tissue. More recently, Relation announced MORGAN, which it describes as a flagship cellular foundation model intended to predict how human cells respond to genetic and pharmacological interventions across disease contexts; it is trained on large-scale multi-omic perturbation data generated from human disease-relevant cellular systems. Strategically, Relation is building a platform-and-data flywheel supported by venture financing and research collaborations with major pharma partners. It has announced multi-program collaborations with Novartis (atopic diseases) and with GSK (fibrotic diseases and osteoarthritis; later expanded with a focus on generating large-scale cellular perturbation datasets and deploying them into MORGAN-related models). The company’s public pipeline messaging emphasizes bone disease starting with osteoporosis and describes Osteomics as a core discovery asset; it also states applicability across immunology and metabolic disease.