Models & datasets

Aura-tahoe-100M

Explore single-cell dataset Tahoe-100M for cancer research.

Tahoe-100M is a giga-scale single-cell perturbation atlas consisting of over 100 million transcriptomic profiles from 50 cancer cell lines exposed to 1,100 small-molecule perturbations. Generated using Vevo Therapeutics' Mosaic high-throughput platform, Tahoe-100M enables deep, context-aware exploration of gene function, cellular states, and drug responses at unprecedented scale and resolution. This dataset is designed to power the development of next-generation AI models of cell biology, offering broad applications across systems biology, drug discovery, and precision medicine.

Reference: https://huggingface.co/datasets/tahoebio/Tahoe-100M

Aura-X-assistant

For rapid structure analysis and function prediction

Model/Tool	Description
Alphafold 2	Protein folding prediction using deep learning
Colabfold	Lightweight, faster version of AlphaFold for structure prediction
ESM	Evolutionary scale modeling for protein structure/folding
Bolz-1, Boltz-2	Protein folding and interaction prediction with affinity
ProTrek	Protein sequence generation and optimization
ESM2	Next-generation ESM model for sequence design and property prediction
Denovo-Pinal	De novo protein sequence generation
Evolla + Llama3 8b	Hybrid model for property prediction and generation
SaProt	Functional annotation and sequence optimization
BiomedNLP	Biomedical natural language processing for contextual predictions
SignalP 6.0	Signal peptide detection
AMP-Net	Antimicrobial peptide prediction
DeepLoc	Protein subcellular localization prediction

Aura-Z-thinking

For advanced modeling and deeper exploration

Model/Tool	Description
OpenMM	Molecular dynamics simulations for protein-ligand systems
PDB Search	Protein Data Bank search for small molecules and proteins
UniProt	Protein sequence and functional information database

Aura-TX-gemma

Tailored for therapeutic discovery and clinical applications

Model/Tool	Description
Wikipedia, Web and PubMed Search	Literature search for translational science
BLASTp	Protein sequence alignment and homology search
Antibody affinity	Prediction of antibody-antigen binding affinity
Molecule Search	Chemical molecule and ligand database search
SMILES to Therapy	Converts SMILES strings into therapeutic candidates
SMILES Description	Generates properties and annotations from SMILES strings
Drug-target interaction	Predicts interactions between drugs and targets
Protein-protein interaction	Predicts interactions between proteins
Toxicity	Predicts potential toxic effects of molecules
Pharmacokinetics	Predicts ADME (absorption, distribution, metabolism, excretion) properties
Developability	Assesses therapeutic molecule manufacturability
CRISPR Repair	Predicts and designs CRISPR-based gene editing outcomes