mims-harvard

Democratizing AI scientists with ToolUniverse

Therapeutics Commons (TDC): Multimodal Foundation for Therapeutic Science

Precision Medicine Knowledge Graph (PrimeKG)

AutoScientists: Self-Organizing Agent Teams for Long-Running Scientific Experimentation

TxAgent: An AI Agent for Therapeutic Reasoning Across a Universe of Tools

A unified multi-task time series model.

Nimfa: Nonnegative matrix factorization in Python

Self-supervised contrastive learning for time series via time-frequency consistency

Graph convolutional neural network for multirelational link prediction

TxGNN: Zero-shot prediction of therapeutic use with geometric deep learning and clinician centered design

Tutorials for Machine Learning on Graphs

Learning Universal Representations of Intermolecular Interactions with ATOMICA

Graph Neural Networks for Irregular Time Series

GraphXAI: Resource to support the development and evaluation of GNN explainers

Subgraph Neural Networks (NeurIPS 2020)

A modern multimodal knowledge graph with type-specific metadata across biomedical domains.

scikit-fusion: Data fusion via collective latent factor models

Domain Adaptation for Time Series Under Feature and Label Shifts

Medea: An AI agent for therapeutic reasoning across biological contexts

CUREBench @ NeurIPS 2025: Benchmarking AI reasoning for therapeutic decision-making at scale

Combinatorial prediction of therapeutic perturbations using causally-inspired neural networks

Graph meta learning via local subgraphs (NeurIPS 2020)

Generalizable AI predicts immunotherapy outcomes across cancers and treatments

Contextual AI models for single-cell protein biology

ClinVec: Unified Embeddings of Clinical Codes Enable Knowledge-Grounded AI in Medicine

OhmNet: Representation learning in multi-layer graphs

SHEPHERD: Few shot learning for phenotype-driven diagnosis of patients with rare genetic diseases

Defending graph neural networks against adversarial attacks (NeurIPS 2020)

ATHENA-R1: AI agent for treatment reasoning over a biomedical tool universe

ProCyon: A multimodal foundation model for protein phenotypes

Time series explainability via self-supervised model behavior consistency

KGARevion: AI Agent for Knowledge-Intensive Biomedical QA

General Strategy for Unlearning in Graph Neural Networks