scFoundation primary supplement, Table4

What this source contributed

Records in this release that cite this source. A record can cite several sources.

• 7 pipelines
• 1 benchmark
• 2 protocols
• 2 dataset subsets
• 14 evaluations
• 14 results
• 10 evidence claims

Results

Values as the source printed them. Compare results only under compatible protocols, datasets and metrics.

Claims

Statements this source supports, each with where in the source it was checked.

• ACTINN: description: Description Cell-type classifier for single-cell RNA-seq that trains a neural network with three hidden layers on datasets with predefined cell types and predicts cell types in other datasets. (qa-descriptions-20261011-b-source-actinn: abstractText; coverage-source-scfoundation-supp: Supplementary Table 4 (cell type annotation on Zheng68K and Segerstolpe))
• CellTypist: description: Description Machine-learning tool for rapid cell-type annotation of single-cell RNA-seq data. (qa-descriptions-20261011-b-source-celltypist: abstractText ('CellTypist, a machine learning tool for rapid and precise cell type annotation'); coverage-source-scfoundation-supp: Supplementary Table 4)
• Scanpy: description: Description Python toolkit for single-cell gene expression analysis, with methods for preprocessing, visualisation, clustering, trajectory inference and differential expression testing. (qa-descriptions-20261011-b-source-scanpy: abstractText; coverage-source-scfoundation-supp: Supplementary Table 4)
• scANVI: description: Description Semi-supervised variant of scVI, a deep generative model of single-cell transcriptomics, that uses existing cell-state annotations to label cells in new datasets. (qa-descriptions-20261011-b-source-scanvi: abstractText ('scANVI, a semi-supervised variant of scVI designed to leverage existing cell state annotations'); coverage-source-scfoundation-supp: Supplementary Table 4)
• scBERT: description: Description Transformer model, pretrained on unlabelled single-cell RNA-seq data in the manner of BERT, that is fine-tuned to annotate cell types. (qa-descriptions-20261011-b-source-scbert: abstractText; coverage-source-scfoundation-supp: Supplementary Table 4)
• scFoundation cell-type annotation comparison Segerstolpe: part of model-coverage-scfoundation-cell-annotation: Part of scFoundation cell-type annotation comparison (Supplementary Table4, PDF p16; Supplementary Note3, Macro F1 score.)
• scFoundation cell-type annotation comparison Zheng68K: part of model-coverage-scfoundation-cell-annotation: Part of scFoundation cell-type annotation comparison (Supplementary Table4, PDF p16; Supplementary Note3, Macro F1 score.)
• scFoundation: description: Description Cell-type annotation with scFoundation, a model that produces contextual cell and gene representations from gene-expression measurements. (catalog-model-scfoundation profile summary (evidence-official-9863509ad10b824dc96b: Results pre-training framework; Methods Embedding, Encoder, Decoder); coverage-source-scfoundation-supp: Supplementary Table 4)
• scFoundation: uses model catalog-model-scfoundation: Uses model scFoundation (Supplementary Table4, PDF p16, row scFoundation)
• SingleCellNet: description: Description Tool that classifies query single-cell RNA-seq data by comparison with reference single-cell RNA-seq data, across platforms and species. (qa-descriptions-20261011-b-source-singlecellnet: title and abstractText; coverage-source-scfoundation-supp: Supplementary Table 4)

Models, benchmarks and other records

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.