中文

Atlases and data resources for virtual cells: platforms, scale and verified data licences

By GlioState Editorial Team · Published · Last updated

Review pending Version 0.1.0 · status: draft · last expert review: not yet reviewed

Layer: General layer · Tissue scope: Pan-tissue

Research question

  • Which general single-cell and perturbation data platforms are in the library, and under what data licence is each available?
  • What do the records say about scale, curation method and access?

Current understanding

Our reading of the library records and our licence check (2026-10-16): the data resources in the library are mostly open (CC0 or CC BY 4.0), with one free-to-access exception under CC BY-NC-SA 4.0. Records describe CELLxGENE Discover (100M+ cells) and scBaseCount (agent-curated, auto-updated) as general single-cell resources. The licence is the licence of the data as shown on the repository page, not that of the paper. Certainty: high for licences as checked on the dates given; scale as stated in the records.

Scope and how to read this page. Scope: library records only (title, venue, one-line summary and verification source); we have not read the full texts for this page, and every statement below is limited to what those records state. Preprints are marked and are not peer reviewed. This page makes no claim about glioma or other tumours. Licence statements come from the repository pages listed in our dataset-licence report and were checked on 2026-10-16; licences can change, so check the source before reuse. VC0043 VC0040

General single-cell platforms. CZ CELLxGENE Discover is recorded as a 100M+ cell single-cell data platform (Nucleic Acids Research 2025); its record notes that it contains glioma datasets. Published CELLxGENE data are CC BY 4.0 under the platform's published-data policy (no per-collection licence field was visible). scBaseCount is recorded as an AI-agent-curated, standardised, auto-updated single-cell repository from Arc (Cell 2026); the CC0 1.0 statement is a page-level statement on Arc's Atlas page, and the data are in a requester-pays bucket with a free quota. Certainty: high that the records and licence pages state this. VC0043 VC0044

Perturbation resources and their data licences. Replogle et al. processed Perturb-seq data: CC BY 4.0 (Figshare+). Tahoe-100M: CC0 1.0 (Hugging Face; the preprint and Cell records describe the same data). X-Atlas/Orion: CC BY-NC-SA 4.0 (free to access, not an open licence; the schema.org licence field is therefore not emitted on its library page). See the large-perturbation-datasets page for what each record says about scale and system. VC0040 VC0041 · preprint VC0042 VC0045 · preprint

Platform and initiative pages. The CZI Virtual Cells Platform is recorded as a hub hosting virtual cell models (organisation page; no publication date on the page), and the Arc Virtual Cell Initiative as a programme combining Perturb-seq data generation, the State model and the Challenge (organisation page). These are organisation descriptions, not evaluations. VC0100 VC0103

Brain-tumor data in these resources (a factual pointer, not a transfer claim). The CELLxGENE record notes glioma datasets, and GBmap (a single-cell and spatial atlas of IDH-wild-type glioblastoma, VC0060) is distributed on CELLxGENE with a verified CC BY 4.0 data licence. This says only where data can be found; it is not evidence that any model on this site's pages works in glioma, and we do not add a transferability assessment here. VC0043 VC0060

Evidence map

Primary studies & datasets

  • VC0043 CZ CELLxGENE Discover: a single-cell data platform for scalable exploration, analysis and modeling of aggregated data. — CZI Cell Science Program et al., Nucleic acids research, 2025
  • VC0044 scBaseCount: An AI agent-curated, standardized, auto-updated single-cell data repository. — Youngblut ND et al., Cell, 2026
  • VC0040 Mapping information-rich genotype-phenotype landscapes with genome-scale Perturb-seq. — Replogle JM et al., Cell, 2022
  • VC0041 · preprint Tahoe-100M : A Giga-Scale Single-Cell Perturbation Atlas for Context-Dependent Gene Function and Cellular Modeling — Zhang J et al., bioRxiv, 2025
  • VC0042 Tahoe-100M: Mapping drug-induced molecular phenotypes at single-cell resolution. — Zhang J et al., Cell, 2026
  • VC0045 · preprint X-Atlas/Orion: Genome-wide Perturb-seq Datasets via a Scalable Fix-Cryopreserve Platform for Training Dose-Dependent Biological Foundation Models — Huang AC et al., bioRxiv, 2025

Related models & tools

  • VC0100 CZI Virtual Cells Platform — Chan Zuckerberg Initiative, web
  • VC0103 Arc Institute Virtual Cell Initiative — Arc Institute, web
  • VC0060 Charting the single-cell and spatial landscape of IDH-wild-type glioblastoma with GBmap. — Ruiz-Moreno C et al., Neuro-oncology, 2025

9 library items cited on this page; 2 are preprints (not peer reviewed).

Entities on this page

What remains uncertain

Licence verification is a point-in-time check of repository pages (2026-10-16). The CELLxGENE per-collection licence field was not visible, and the scBaseCount CC0 statement is page-level rather than a licence file in its code repository. VC0043 VC0044

Scale figures are those stated in the records (for example 100M+ cells); we did not independently count cells or check overlap between resources, so resources may contain overlapping underlying datasets. VC0043 VC0041 · preprint

For research and education only. Not medical advice and not a recommendation of any diagnosis, treatment or product. Patients should consult their own physicians.