Configuration
scGPT, randomly initialised (Kedzierska et al. 2025)
scGPT, randomly initialised: scGPT, randomly initialised cell representation; No pretraining (random initialisation); zero-shot; scGPT cell embedding.
14 evaluations · 97 results4 open disagreements between studies involve its model
Overview
An explanatory profile has not yet been reviewed for this record.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluations and results
14 evaluations · 97 results. Different protocols are not a single leaderboard.
Sorted by Adjusted rand index (Louvain clusters at best-NMI resolution versus cell-type labels; whole dataset; full embedding dimensions) (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.
| Tested configuration | Protocol and dataset | Adjusted rand index (Louvain clusters at best-NMI resolution versus cell-type labels; whole dataset; full embedding dimensions) | Adjusted rand index (Louvain clusters at best-NMI resolution versus cell-type labels; whole dataset; UMAP projection of the embedding) | Avg bio (AvgBIO (mean of cell-type ASW, NMI and ARI; NMI and ARI on Louvain clusters at the best-NMI resolution); whole dataset; full embedding dimensions) | Avg bio (AvgBIO (mean of cell-type ASW, NMI and ARI; NMI and ARI on Louvain clusters at the best-NMI resolution); whole dataset; UMAP projection of the embedding) | Normalized mutual information (Louvain clusters at best-NMI resolution versus cell-type labels; whole dataset; full embedding dimensions) | Normalized mutual information (Louvain clusters at best-NMI resolution versus cell-type labels; whole dataset; UMAP projection of the embedding) | Scib asw label (cell-type ASW; whole dataset; full embedding dimensions) | Scib asw label (cell-type ASW; whole dataset; UMAP projection of the embedding) | Average batch score (average batch score (mean of batch integration score and PCR batch score); whole dataset; full embedding dimensions) | Average batch score (average batch score (mean of batch integration score and PCR batch score); whole dataset; UMAP projection of the embedding) | Rank (rank among the eight representations by mean biological-conservation and batch-mixing scores on this dataset (1 = best); full embedding dimensions) | Rank (rank among the eight representations by mean biological-conservation and batch-mixing scores on this dataset (1 = best); UMAP projection of the embedding) | Scib asw batch (batch integration score (batch ASW per cell type); whole dataset; full embedding dimensions) | Scib asw batch (batch integration score (batch ASW per cell type); whole dataset; UMAP projection of the embedding) | Scib pcr batch (principal component regression on batch; whole dataset; full embedding dimensions) | Scib pcr batch (principal component regression on batch; whole dataset; UMAP projection of the embedding) | Rank (overall average ranking across datasets, based on mean biological-conservation and batch-mixing scores (1 = best); full embedding dimensions) | Rank (overall average ranking across datasets, based on mean biological-conservation and batch-mixing scores (1 = best); UMAP projection of the embedding) | Rank (rank of the median ARI score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median ASW score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median Average batch score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median Average BIO score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median Batch integration score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median NMI score across datasets among the eight representations (1 = best); full embedding dimensions) | Rank (rank of the median PCR score across datasets among the eight representations (1 = best); full embedding dimensions) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, Immune (330k) (Kedzierska et al. 2025) | 0.31 (best in column) | Not reported | 0.45 (best in column) | Not reported | 0.54 (best in column) | Not reported | 0.49 | Not reported | 0.91 | Not reported | 2 (best in column) | Not reported | 0.88 | Not reported | 0.95 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, PBMC (95k) (Kedzierska et al. 2025) | 0.18 | Not reported | 0.38 | Not reported | 0.47 | Not reported | 0.5 (best in column) | Not reported | Not reported | Not reported | 8 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, PBMC (12k) (Kedzierska et al. 2025) | 0.1 | Not reported | 0.22 | Not reported | 0.1 | Not reported | 0.45 | Not reported | 0.84 | Not reported | 6 | Not reported | 0.97 (best in column) | Not reported | 0.71 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, Tabula Sapiens (483k) (Kedzierska et al. 2025) | 0.09 | Not reported | 0.32 | Not reported | 0.44 | Not reported | 0.44 | Not reported | 0.9 | Not reported | 3 | Not reported | 0.89 | Not reported | 0.91 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, Pancreas (16k) (Kedzierska et al. 2025) | 0 | Not reported | 0.17 | Not reported | 0.01 | Not reported | 0.49 | Not reported | 0.94 (best in column) | Not reported | 4 | Not reported | 0.96 | Not reported | 0.93 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, Pancreas (8k) (Kedzierska et al. 2025) | 0 | Not reported | 0.17 | Not reported | 0.02 | Not reported | 0.48 | Not reported | 0.94 (best in column) | Not reported | Not reported | Not reported | 0.96 | Not reported | 0.92 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring, full embedding dimensions, Evaluated datasets, aggregated (Kedzierska et al. 2025) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 6 | Not reported | 8 | 7 | 1 | 8 | 2 | 7 | 1 |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, Evaluated datasets, aggregated (Kedzierska et al. 2025) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | 6 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, Immune (330k) (Kedzierska et al. 2025) | Not reported | 0.29 (best in column) | Not reported | 0.41 (best in column) | Not reported | 0.46 (best in column) | Not reported | 0.47 | Not reported | 0.76 | Not reported | 4 | Not reported | 0.56 | Not reported | 0.96 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, Pancreas (16k) (Kedzierska et al. 2025) | Not reported | 0 | Not reported | 0.15 | Not reported | 0.01 | Not reported | 0.44 | Not reported | 0.9 | Not reported | 3 (best in column) | Not reported | 0.82 | Not reported | 0.98 (best in column) | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, Pancreas (8k) (Kedzierska et al. 2025) | Not reported | 0 | Not reported | 0.15 | Not reported | 0.02 | Not reported | 0.43 | Not reported | 0.88 | Not reported | Not reported | Not reported | 0.8 | Not reported | 0.95 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, PBMC (12k) (Kedzierska et al. 2025) | Not reported | 0.06 | Not reported | 0.19 | Not reported | 0.09 | Not reported | 0.43 | Not reported | 0.93 (best in column) | Not reported | 8 | Not reported | 0.96 (best in column) | Not reported | 0.9 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, PBMC (95k) (Kedzierska et al. 2025) | Not reported | 0.27 | Not reported | 0.38 | Not reported | 0.34 | Not reported | 0.53 (best in column) | Not reported | Not reported | Not reported | 4 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
| scGPT, randomly initialised (Kedzierska et al. 2025)Evaluation | Kedzierska et al. 2025 cell embedding scoring after UMAP projection, Tabula Sapiens (483k) (Kedzierska et al. 2025) | Not reported | 0.22 | Not reported | 0.32 | Not reported | 0.4 | Not reported | 0.32 | Not reported | 0.8 | Not reported | 6 | Not reported | 0.69 | Not reported | 0.91 | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported | Not reported |
All 97 result rows with coverage, uncertainty and sources
Filter evaluations
Applied filters: All linked evaluations
Source checking is not independent reproduction. Release 2026-10-23-5347695500ed.
Use this configuration
How it works, versions and access
Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
0 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|
No evidence rows match these filters. Choose another scope or clear the search.
Sources and history
Release 2026-10-23-5347695500ed · Record review: source checked
1 source record and release history
- Zero-shot evaluation reveals limitations of single-cell foundation models (Kedzierska et al. 2025, Genome Biology 26:101) · Original source · Version of record, Genome Biology 26:101, published 2025-04-18
Technical metadata and extraction receipts
Stable ID: gfm-guides-20261014-kedzierska2025-config-scgpt-random
- areas
- single-cell
- contexts
- research
- method types
- foundation_model
- reported name
- scGPT random
- foundation model eligible
- true
- training setting
- No pretraining (random initialisation); zero-shot
- representation
- scGPT cell embedding
- version
- scGPT code 0.1.6 (2023-09-14)
- checkpoint
- Randomly initialised scGPT (no pretrained weights)
- access reuse
- Evaluation code: github.com/microsoft/zero-shot-scfoundation (MIT), archived on Zenodo; not pinned to a commit in the article
- source locator
- Pages 7-8 'Models and baselines'; page 11 'Code availability'
- provenance note
- scGPT, randomly initialised cell representation as evaluated zero-shot by Kedzierska et al. 2025.
Related records
- configuration of: scGPT
- system: scGPT random cell embedding on evaluated datasets (Kedzierska et al. 2025)
- system: scGPT random cell embedding on Immune (330k) (Kedzierska et al. 2025)
- system: scGPT random cell embedding on Pancreas (16k) (Kedzierska et al. 2025)
- system: scGPT random cell embedding on Pancreas (8k) (Kedzierska et al. 2025)
- system: scGPT random cell embedding on PBMC (12k) (Kedzierska et al. 2025)
- system: scGPT random cell embedding on PBMC (95k) (Kedzierska et al. 2025)
- system: scGPT random cell embedding on Tabula Sapiens (483k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on evaluated datasets (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on Immune (330k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on Pancreas (16k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on Pancreas (8k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on PBMC (12k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on PBMC (95k) (Kedzierska et al. 2025)
- system: scGPT random UMAP-projected embedding on Tabula Sapiens (483k) (Kedzierska et al. 2025)