Afficher les postes ouverts en Français

Predict. Explain. Discover. Valence Labs is Recursion’s AI research engine. Leveraging the full power of Recursion’s platform, data, and computing infrastructure, we develop new ways to predict, explain, and ultimately decode biology. Learn more Our focus Predict Building on more than a decade of experience in perturbative biology at Recursion—from scaled phenomics and transcriptomics datasets to multimodal foundation models—we are modeling the functional response of cells to perturbation at unprecedented scale. Explain Perturbations rewire cellular function by altering molecular interactions—binding, signaling, dynamics, and more. By combining interventional data with novel methods for predicting and simulating molecular interactions at scale, we are generating casual explanations for how molecular interventions shape cellular function. Discover Using lab-in-the-loop engines of biological discovery, we bridge functional readouts with mechanistic understanding to generate, test, and refine novel therapeutic hypotheses to accelerate and improve drug discovery outcomes. Powered by   A unique set of industry-leading ingredients Unprecedented Data Generation Recursion’s OS houses automated biology and chemistry labs capable of generating enormous interventional datasets, today spanning > 60 petabytes of data across phenomics, transcriptomics, and other modalities. Massive Computing Power Using Recursion’s BioHive, the pharmaceutical industry’s leading supercomputer, we have the ability to run training and inference workflows at industry-leading scale. World-Class Talent With a mission-driven, interdisciplinary team, equally fluent in computer science and biology, we have the right people in place to realize our ambitious goal of decoding biology to radically improve lives. Our perspective A vision for virtual cells—mechanistic models of cellular function that guide the discovery of novel therapeutics. Read the Perspective Virtual Cells: Predict, Explain, Discover Abstract The objective of drug discovery is to accurately infer the effects of treatments on patients. Drug discovery would therefore be greatly improved if there existed computational models thataccurately predicted the response of patients to interventions, since this would allowpractitioners to safely and economically test and optimize a wide range of therapeutichypotheses before ever starting a human clinical trial. Even a more “modest” model that couldaccurately predict the functional response of a wide variety of cells to genetic and chemicalinterventions would be of tremendous value in designing effective and safe therapeutics morelikely to produce positive outcomes in the clinic. Creating such virtual cells has long been a goalof the computational research community that even today remains an ambition due to thedaunting scale and complexity of the biomolecular interactions mediating cellular function.Nevertheless, a confluence of technological advances suggest that there has never been abetter time to attempt to build virtual cells. In this perspective, we describe Valence Labs’s visionfor virtual cells as a transformative platform to indus drug discovery. We set the contextfor our vision by reviewing historical progress, and outline their integration within a largerframework of agentic systems that continuously refine our mechanistic understanding of humanphysiology. We highlight recent advances in machine learning, computational power, and datageneration that now enable robust simulation of cellular functional responses, and we outlinekey modeling considerations, evaluation benchmarks, and a roadmap for future research. Featured highlights Stay informed Blog May 22nd, 2025 TxPert: Predicting Cellular Responses to Unseen Genetic Perturbations We introduce TxPert: a state-of-the-art model that leverages multiple biological knowledge networks to accurately predict transcriptional responses under OOD scenarios. Read more Blog May 20th, 2025 Advancing Drug Discovery Outcomes with Virtual Cells at Recursion Predict, Explain, Discover: The Pillars of the Virtual Cell Our view of the virtual cell rests on three interconnected capabilities: Read more Blog November 26th, 2024 Introducing OpenQDC – The Open-Source Hub of ML-Ready Quantum Datasets We curated and consolidated 40+ quantum mechanics (QM) datasets, covering 1.5 billion geometries across 70 atom species and 250+ QM… Read more