Open Source Data
Therapeutics Data Commons
TDC is a is an open-source and comprehensive data hub of machine learning datasets for therapeutics. TDC covers a wide range of tasks, including target discovery, activity screening, efficacy, saffety, and manufacturing. TDC contains datasets describing diverse types of products, including small molecules, antibodies, vaccines, and miRNA.
The TDC platform implements numerous data functions for dataset processing, model evaluation, molecule generation oracles, and more.
Stanford Biomedical Network Dataset Collection
BioSNAP is a collection diverse biomedical networks, inclusing protein-protein interaction networks, single-cell similarity networks, drug-drug interaction networks.
BioSNAP datasets contain metadata on graphs and node features, and can be easily linked to external repositories of biological knowledge.
The Open Graph Benchmark
OGB is a collection of benchmark datasets, data loaders, and evaluators for graph machine learning. Datasets cover a variety of graph machine learning tasks and real-world applications.
The OGB data loaders are fully compatible with popular graph deep learning frameworks, including Pytorch Geometric and DGL. They provide automatic dataset downloading, standardized dataset splits, and unified performance evaluation.
Multimodal cancer network
Multimodal network centered on genes frequently mutated in cancer patients
The multimodal cancer network integrates information on chemicals, diseases, molecular functions, genes, and protein.
The dataset has 21 types of biologically meaningful associations (edge types): chemical-chemical, chemical-protein, disease-chemical, disease-disease, disease-function, disease-gene, function-function, gene-gene (split into 6 edge types by interaction type), gene-protein, protein-function, and protein-protein interactions.
The network has 20 K nodes and 3.4 M edges.
Giga-scale biological network
The giga-scale biological network is one of the largest networks ever constructed in biology. The network integrates protein and genetic interaction data from more than two thousand species.
The network has 10 M nodes and 2.3 B edges.
Tree of life
Protein interactomes across the tree of life
The dataset contains protein interactomes from 1,840 species across the tree of life. The dataset contains rich metadata about prrteins, including their homology relationships
The dataset also contains metadata about species, including taxonomy of species, phylogenetic reltionships, and ecological information on environments and habitats in which species live.
Network of drugs, proteins, and side effects
The polypharmacy network is a highly multi-relational network, consisting of protein-protein interactions, drug-protein targets, and drug-drug interactions encoded by polypharmacy side effects.
The network has 20 K nodes and 5 M edges, which are split into 1 K distinct edge types.
Tissue-specific protein dataset
The dataset contains protein-protein interaction networks specific to 107 human tissues, a tissue hierarchy of anatomical relationships between tissues, and tissue-specific gene-function annotations.
Human knowledge network
The human knowledge network contains interactions between proteins, diseases, biological processes, side effects, and drugs.
The network has 98 K nodes and 8 M edges, which are split into 42 distinct types of biologically relevant molecular interactions.
We are excited to co-organize Workshop on AI in Health: Transferring and Integrating Knowledge for Better Health at the Web (WWW) conference. The call for papers is open! We also announce the AI in Health Data Challenge.
We are excited to see published our paper on DeepPurpose, a deep learning library for drug-target interaction prediction, and a paper on skipGNN, a graph neural network for predicting molecular interactions.
Our research won the Bayer Early Excellence in Science Award. We are honored to have received this recognition!
We are thrilled to announce Therapeutics Data Commons (TDC)! We invite you to join TDC. TDC is an open-source and community-driven effort.
On behalf of the NSF, we are organizing the National Symposium on Drug Repurposing for Future Pandemics. We have a stellar lineup of invited speakers! Register at www.drugsymposium.org.
Hot off the press! MARS, an approach for discovering novel cell types across heterogeneous single-cell experiments is just published in Nature Methods.
The MITxHarvard Women in AI initiative talked with Marinka about AI, machine learning, and the role of new technologies in biomedical research.
Tune in to the podcast with Marinka about machine learning to drug development. The discussion focuses on open research questions in the field, including how to limit the search space of high-throughput screens, design drugs entirely from scratch, and identify likely side-effects of combining drugs in novel ways.
We have a new opening for a postdoctoral research fellow in novel machine learning methods to combat COVID-19! Submit your application by September 1, 2020.
DeepPurpose is a deep learning library for drug-target interaction prediction and applications to drug repurposing and virtual screening.
Subgraph neural networks learn powerful subgraph representations that create fundamentally new opportunities for predictions beyond nodes, edges, and entire graphs.
GNNGuard can defend graph neural networks against a variety of training-time attacks. Remarkably, GNNGuard can restore state-of-the-art performance of any GNN in the face of adversarial attacks.
G-Meta is a meta-learning approach for graphs that quickly adapts to new prediction tasks using only a handful of data points. G-Meta works in most challenging, few-shot learning settings and scales to massive interactomics data as we show on our new Networks of Life dataset comprising of 1,840 networks.
A new paper introducing the Open Graph Benchmark, a diverse set of challenging and realistic benchmark datasets for graph machine learning.
Marinka is co-editing a special issue of IEEE Big Data on AI for COVID-19. In light of the urgent need for data-driven solutions to mitigate the COVID-19 pandemic, the special issue will aim for a fast-track peer review.