March 3, 20260 citationsOpen Access

CytoVerse: Single-Cell AI Foundation Models in the Browser

RCRobert CurrieUniversity of California, Santa Cruz AVA. VaniachineNational Research Tomsk State University MMMohammed A. Mostajo-RadjiUniversity of California, Santa Cruz

Key Points

CytoVerse enables single-cell datasets to be processed entirely in the browser, enhancing data privacy.
The framework utilizes ONNX for deploying models, allowing for effective use without server computation.
Compressed indexing techniques allow for efficient searching of over 20 million cell references directly from the client.
The lightweight protocol facilitates safe sharing of embeddings, minimizing exposure of raw data.

Abstract

Mapping single-cell datasets to large atlases is often hindered by server constraints and privacy concerns. We present CytoVerse, a framework that runs scRNA-seq Foundation Models (scFM) entirely in the browser. Three key contributions enable this: (1) deploying models via ONNX without server side compute; (2) using compressed indexing (IVFPQ) to search a more then 20 million cell reference from the client; and (3) a lightweight protocol for sharing embeddings across consortia without exposing raw data. CytoVerse thereby provides a scalable, privacy preserving framework for distributed single-cell analysis.

问 AI

Bookmark

View Full Paper