rewire.it
Model

scFoundation

scFoundation produces contextual cell and gene representations from gene-expression measurements.

Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Results are available for configurations using this model. Their fitted heads, extra inputs and evaluation settings are kept separate below.

1 evaluated configuration using this model

How it worksscFoundation workflow
scFoundation workflow1. Expression and read-depth indicators. Then: 2. Gene and continuous-value embeddings. Then: 3. Sparse-input transformer encoder. Then: 4. Full-gene Performer decoder. Then: 5. Cell/gene embeddings or expressionscFoundation workflow1. Expression and read-depth indicators. Then: 2. Gene and continuous-value embeddings. Then: 3. Sparse-input transformer encoder. Then: 4. Full-gene Performer decoder. Then: 5. Cell/gene embeddings or expressionscFoundation workflow1. Expression and read-depth indicators. Then: 2. Gene and continuous-value embeddings. Then: 3. Sparse-input transformer encoder. Then: 4. Full-gene Performer decoder. Then: 5. Cell/gene embeddings or expression

Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.

Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Overview

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Related configurations, pipelines and services

These configurations, services and pipelines use this model within their own configurations. Their results, where available, are not assigned to the underlying model.

Use this model

How it works, versions and access

How it works

How it works

scFoundation embeds gene identity and continuous expression together with source and target read-depth indicators. Its encoder processes only nonzero, unmasked genes. Those contextual embeddings are combined with zero and mask embeddings before a Performer decoder predicts expression across the full gene vocabulary. Pooled encoder outputs represent cells; decoder outputs provide gene-level context.

Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Versions and reproducibility

scFoundation / xTrimoscFoundation-alpha; the repository exposes separately configured embedding, enhancement and downstream prediction workflows. Fixed input gene vocabulary of 19,264 genes; not a nucleotide token context.

Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Strengths, limitations and unresolved questions

Strengths and limitations

Profile review details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Stable record: catalog-model-scfoundation

Specifications

Inputs, training, access and other details

Explanatory profile: limited source coverage · Automated source review, 2026-09-16. Review applies to the cited claims; unresolved fields are listed below. Numerical results retain their own review status.

Inputs, outputs and configuration
PropertyDescription and evidence
Model typeTranscriptomic representation model
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
ArchitectureAn asymmetric transformer encoder-decoder: learned continuous-expression embeddings enter a transformer encoder for nonzero, unmasked genes, then a Performer decoder reconstructs the full gene set.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
InputsSingle-cell or bulk expression aligned to the documented 19,264-gene vocabulary.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
OutputsCell embeddings, contextual gene embeddings and outputs of separately configured downstream models.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Parameters100 million.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Known versionsscFoundation / xTrimoscFoundation-alpha; the repository exposes separately configured embedding, enhancement and downstream prediction workflows.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Training dataThe June 2023 manuscript describes more than 50M human cells collected from GEO, Single Cell Portal, HCA and EMBL-EBI, aligned to 19,264 genes after quality control.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Training cutoffThe June 2023 manuscript lists GEO, Single Cell Portal, HCA and EMBL-EBI as collection sources; its data-collection section does not give a shared last-included-study date. · Not reported in inspected sources
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Context limitsFixed input gene vocabulary of 19,264 genes; not a nucleotide token context.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Weights licenceSeparate Model License; the Apache source-code notice explicitly excludes model-weight rights.
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
AccessOfficial project documentation and implementation: https://github.com/biomap-research/scFoundation
Sources (3)biomap-research/scFoundation: README.md; biomap-research/scFoundation: model/README.md; scfoundation-preprint: Primary paper PDF · June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README
Code licenceApache-2.0
Sourcesbiomap-research/scFoundation: LICENSE · LICENSE: licence text

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.

58 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-23-2b89723c6dd9
Property and statementOriginal source and locationReview and provenance
Diagram caption
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Individual claims
biomap-research/scFoundation: model/README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: f62879233ecf5fac407cd516aca2ddb3df20370523cb111c9d057676b4721d5d

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram caption
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Individual claims
scfoundation-preprint: Primary paper PDF

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: bioRxiv manuscript posted 2023-06-15
Retrieved: 2026-09-16T20:35:33.635910+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: e2ce9d623a09b53eb863ae6ca359fe2956b7b0a7c3ebe7de6911aca3a660f737

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram caption
Conceptual summary of the documented data flow; optional inputs and configured downstream stages must be reported for a reproducible evaluation.
Individual claims
biomap-research/scFoundation: README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.caption

Source artifact SHA-256: 02a7ae44cf2cc9948b5c1520b18a261b45abf9f17e3149e37d9f48072184f376

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram steps
  • Expression and read-depth indicators
  • Gene and continuous-value embeddings
  • Sparse-input transformer encoder
  • Full-gene Performer decoder
  • Cell/gene embeddings or expression
Individual claims
biomap-research/scFoundation: model/README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: f62879233ecf5fac407cd516aca2ddb3df20370523cb111c9d057676b4721d5d

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram steps
  • Expression and read-depth indicators
  • Gene and continuous-value embeddings
  • Sparse-input transformer encoder
  • Full-gene Performer decoder
  • Cell/gene embeddings or expression
Individual claims
scfoundation-preprint: Primary paper PDF

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: bioRxiv manuscript posted 2023-06-15
Retrieved: 2026-09-16T20:35:33.635910+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: e2ce9d623a09b53eb863ae6ca359fe2956b7b0a7c3ebe7de6911aca3a660f737

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram steps
  • Expression and read-depth indicators
  • Gene and continuous-value embeddings
  • Sparse-input transformer encoder
  • Full-gene Performer decoder
  • Cell/gene embeddings or expression
Individual claims
biomap-research/scFoundation: README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.steps

Source artifact SHA-256: 02a7ae44cf2cc9948b5c1520b18a261b45abf9f17e3149e37d9f48072184f376

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram title
scFoundation workflow
Individual claims
biomap-research/scFoundation: model/README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.title

Source artifact SHA-256: f62879233ecf5fac407cd516aca2ddb3df20370523cb111c9d057676b4721d5d

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram title
scFoundation workflow
Individual claims
scfoundation-preprint: Primary paper PDF

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: bioRxiv manuscript posted 2023-06-15
Retrieved: 2026-09-16T20:35:33.635910+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.title

Source artifact SHA-256: e2ce9d623a09b53eb863ae6ca359fe2956b7b0a7c3ebe7de6911aca3a660f737

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Diagram title
scFoundation workflow
Individual claims
biomap-research/scFoundation: README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.diagram.title

Source artifact SHA-256: 02a7ae44cf2cc9948b5c1520b18a261b45abf9f17e3149e37d9f48072184f376

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Model type
Transcriptomic representation model
Individual claims
biomap-research/scFoundation: model/README.md

Original source ↗

June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 397631c495eddf9ad6644fc00c6ea8139e651245
Retrieved: 2026-09-16T19:46:18.581063+00:00

source checked

automated source review · 2026-09-16

Audit details

Inspected pinned official documentation, relevant implementation files and named primary-paper sections. Claims are limited to those artifacts. Remaining field extraction and identity conflicts are explicit; no new performance claims, model runs or human review are implied.

Field: attributes.profile.facts.0.value

Source artifact SHA-256: f62879233ecf5fac407cd516aca2ddb3df20370523cb111c9d057676b4721d5d

Hash scope: SHA-256 of retrieved original artifact bytes

Format: original_artifact

Inspected artifact

Sources and history

View linked audit checks and correction history

Release 2026-09-23-2b89723c6dd9 · Record review: discovered

5 source records and release historyDownload this release
Technical metadata and extraction receipts

Stable ID: catalog-model-scfoundation

areas
cells-tissues
method types
foundation model
entity level
family
version
100M
reported name
scFoundation
access
Public code; model weights have separate terms that must be checked.
method type
foundation model
historical missing metadata
checkpoint revision: not_yet_extracted; training data: not_yet_extracted; licence: not_yet_extracted
metadata review scope
historical_missing_metadata preserves the original discovery state. Current descriptive evidence and missingness are recorded in profile.facts; numerical-result review is separate.
entity classification
review date: 2026-09-17; rationale: The cited profile describes a named learned biological predictor or representation model/family. Preserve this identity separately from task-specific fitting, individual checkpoints, pipelines and hosted access.; source ids: evidence-official-bfbc8babf3630ddb3ed7; evidence-official-7972ce2bdd6a5e40d8ad; evidence-official-9863509ad10b824dc96b; source locator: June 15 2023 scFoundation manuscript: Results pre-training framework; Methods Data collection, Embedding, Encoder, Decoder and Read-depth-aware pre-training; official model README; ambiguities: None recorded
Related records

Suggest a correction