TAAFT
Free mode
100% free
Freemium
Free Trial
Deals

Rhaister

perturbation combinations. It works directly on summary statistics from standard single-cell analysis pipelines, including log2 fold change, Mann-Whitney U-test p-values, and expression deltas, which lets it train in seconds and predict in milliseconds. The few-shot version predicts unseen context-perturbation combinations using additive baseline modeling, ridge regression, and a calibration network, while Rhaister-O supports zero-shot prediction for entirely unseen cell lines using only baseline expression. It is designed for biological perturbation modeling, drug-response prediction, cytokine response prediction, CRISPR Perturb-seq analysis, and fast scientific workflows where observed statistics are sufficient for accurate response prediction.
New Structured_data Gen 7
Released: June 10, 2026

Overview

Rhaister is Tahoe Bio’s open model for predicting cellular drug and perturbation responses from aggregated single-cell summary statistics instead of raw single-cell data.

About Tahoe Therapeutics

Tahoe Therapeutics builds gigascale single-cell perturbation maps and AI virtual cell models to accelerate drug discovery. Its Mosaic platform generates hundreds of millions of single-cell data points measuring how drug molecules interact with patient cells, with applications in oncology and precision medicine.

Industry: Biotechnology
Company Size: 25
Location: South San Francisco, California, US
Website: tahoebio.ai
View Company Profile
Last updated: June 18, 2026
0 AIs selected
Clear selection
#
Name
Task