ATHENA
ATHENA-R1 is an AI agent for treatment reasoning over a biomedical tool universe
PROTON
Graph AI Generates Neurological Hypotheses Validated in Molecular, Organoid, and Clinical Systems
StratCP
Error Controlled Decisions for Safe Use of Medical Foundation Models
COMPASS
A Foundation Model for Predicting Immunotherapy Outcomes Across Cancers and Treatments
ATOMICA
Universal Geometric AI for Molecular Interactions across Biomolecular Modalities
Magneton
Greater than the Sum of Its Parts - Building Substructure into Protein Encoding Models
TxAgent
TxAgent is an AI agent for therapeutic reasoning across a universe of tools.
CLEF
CLEF is a controllable sequence editing model for counterfactual reasoning about both immediate and delayed effects.
ProCyon
ProCyon is a groundbreaking foundation model for modeling, generating, and predicting protein phenotypes.
ClinVec and ClinGraph
ClinVec - Unified Embeddings of Clinical Codes Enable Knowledge-Grounded AI in Medicine
Madrigal
Multimodal AI Predicts Clinical Outcomes of Drug Combinations from Preclinical Data
KGARevion
KGARevion - An AI Agent for Knowledge-Intensive Biomedical QA
PocketFlow
Generalized Protein Pocket Generation with Prior-Informed Flow Matching
Therapeutics Data Commons 2.0
Multimodal Foundation for Therapeutic Science
SPECTRA
Evaluating Generalizability of Artificial Intelligence Models for Molecular Datasets
PDGrapher
Combinatorial Prediction of Therapeutic Perturbations Using Causally-Inspired Neural Networks
TxGNN
A Foundation Model for Clinician Centered Drug Repurposing
TimeX
Encoding Time-Series Explanations through Self-Supervised Model Behavior Consistency
Raincoat
Domain Adaptation for Time Series Under Feature and Label Shifts
SHEPHERD
Deep Learning for Diagnosing Patients with Rare Genetic Diseases
metapaths
Similarity Search in Heterogeneous Knowledge Graphs via Meta Paths
Mutual Interactors
Phenotype Discovery in Molecular Interaction Networks
Raindrop
Graph-Guided Network for Irregularly Sampled Multivariate Time Series
Therapeutics Data Commons
Machine Learning Datasets and Tasks for Drug Discovery and Development
DeepPurpose
Deep learning library for drug-target interaction prediction and applications to drug repurposing and virtual screening
Graph Query Embeddings
Method for embedding logical queries on knowledge graphs
Network-Guided Matrix Completion
Method for probabilistic prediction and imputation of interactions using prior knowledge
Multi-BioNER
Deep multi-task learning for cross-type biomedical named entity recognition
Latest News
Jul 2026: An AI agent for therapeutic reasoning across biological contexts
Medea is an AI agent that nominates therapeutic targets across cell type contexts, predicts synthetic lethality in cancer cell lines, and forecasts immunotherapy response from multimodal patient profiles. To test whether Medea can identify the genetic vulnerabilities that DNA-damaging treatments exploit, we use it to scan 238,046 gene combinations for synthetic lethality. [Project website]
Jul 2026: Immune Checkpoint Inhibitors in Nature Medicine
COMPASS is a pan-cancer foundation model that predicts immunotherapy response from tumor microenvironments and highlights the biology driving that response. [Nature Medicine paper] [Harvard Medicine News]
Jul 2026: ATHENA Agent for Treatment Reasoning
Treatment reasoning underpins every therapeutic decision in medicine. ATHENA an AI agent for treatment reasoning across all FDA approved drugs since 1939, trained by reinforcement learning over a universe of 212 biomedical tools. [Project website]
Jun 2026: MedLog
MedLog is an open protocol for event-level logging of medical AI, validated across four real-world pilots in the US, Switzerland, and Vietnam to enable auditing, monitoring, and governance of AI systems. [Paper] [Project website]
Jun 2026: Biological Reasoning Models
Biological reasoning models combine large language models with models of biological data, including DNA, RNA, and proteins. New preprint on training and evaluating 100+ biological reasoning models.
May 2026: AutoScientists
May 2026: Three papers Accepted at ICML 2026
May 2026: Agentic AI for Science in Nature Methods
Our work on agentic AI for science featured in Nature Methods. This piece features: (1) ToolUniverse — an open platform enabling AI agents to use scientific tools and databases at scale, (2) ClawInstitute — shared research boards for long-running collaborative discovery where agents co-develop ideas over time, and (3) Medea — an omics AI agent for large-scale biological reasoning and analysis.
Apr 2026: OptimusKG: A Modern Knowledge Graph
OptimusKG brings biomedical knowledge into a modern multimodal knowledge graph. It supports graph AI, knowledge-grounded retrieval with large language models, and discovery workflows that generate and evaluate biomedical hypotheses.
Apr 2026: Relational Reasoning of LLMs
Apr 2026: ARK Accepted at ACL 2026
ARK is an AI agent for autonomous knowledge graph exploration with adaptive breadth-depth retrieval. [ARK Agent CLI]
Mar 2026: Open 'AI Scientists' Initiative
Excited to launch Open AI Scientists, our initiative to empower scientific discovery with AI scientists. [https://www.openscientist.ai]
Mar 2026: Generalist Biological AI in Nature Biotechnology
Review on Generalist biological artificial intelligence in modeling the language of life in Nature Biotechnology.
Mar 2026: Claw Institute
Claw Institute is a research exchange for AI scientists. It gives agents a shared space to publish ideas, challenge claims, use scientific tools, and build on one another’s work. These early interactions point to a new mode of discovery in which societies of AI scientists participate in discovery loops alongside human researchers.
Mar 2026: Five Papers Accepted at ICLR 2026
Feb 2026: Overton Prize
Our research has been recognized with the 2026 Overton Prize.
Feb 2026: Foundation Models that Can 'Act or Defer'
New preprint on the safe use of medical foundation models for real-world decision making. [Project website]
Feb 2026: Reasoning Model for Longitudinal Data
New preprint on multimodal reasoning model for time series.
Feb 2026: DNA-Conditioned Models of Single-Cell Perturbations
Feb 2026: Context Switching AI in Nature Medicine
Tweets