Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models
bioRxiv preprint version 1, posted 2026-05-20 · Preprint
What this source contributed
Records in this release that cite this source. A record can cite several sources.
Results
Values as the source printed them. Compare results only under compatible protocols, datasets and metrics.
Models, benchmarks and other records
Models (1)
Methods (1)
Configurations (21)
- DeepSEM (Ye et al. 2026 baseline)
- Geneformer attn GRN (Ye et al. 2026)
- Geneformer coshid GRN (Ye et al. 2026)
- Geneformer costok GRN (Ye et al. 2026)
- GENIE3 (Ye et al. 2026 baseline)
- GRNBoost (Ye et al. 2026 baseline)
- LangCell attn GRN (Ye et al. 2026)
- LangCell coshid GRN (Ye et al. 2026)
- LangCell costok GRN (Ye et al. 2026)
- scCello attn GRN (Ye et al. 2026)
- scCello coshid GRN (Ye et al. 2026)
- scCello costok GRN (Ye et al. 2026)
- scFoundation attn GRN (Ye et al. 2026)
- scFoundation coshid GRN (Ye et al. 2026)
- scFoundation costok GRN (Ye et al. 2026)
- scGPT attn GRN (Ye et al. 2026)
- scGPT coshid GRN (Ye et al. 2026)
- scGPT costok GRN (Ye et al. 2026)
- scPRINT attn GRN (Ye et al. 2026)
- scPRINT coshid GRN (Ye et al. 2026)
- scPRINT costok GRN (Ye et al. 2026)
Benchmarks (1)
Protocols (8)
- Ye et al. 2026 GRN reconstruction, TFs+500, cell-type-specific ChIP-seq reference
- Ye et al. 2026 GRN reconstruction, TFs+500, non-specific ChIP-seq reference
- Ye et al. 2026 GRN reconstruction, TFs+500, OmniPath reference
- Ye et al. 2026 GRN reconstruction, TFs+500, STRING reference
- Ye et al. 2026 GRN reconstruction, TFs+1000, cell-type-specific ChIP-seq reference
- Ye et al. 2026 GRN reconstruction, TFs+1000, non-specific ChIP-seq reference
- Ye et al. 2026 GRN reconstruction, TFs+1000, OmniPath reference
- Ye et al. 2026 GRN reconstruction, TFs+1000, STRING reference
Dataset subsets (14)
- BEELINE hESC, TFs+500 (Ye et al. 2026)
- BEELINE hESC, TFs+1000 (Ye et al. 2026)
- BEELINE hHep, TFs+500 (Ye et al. 2026)
- BEELINE hHep, TFs+1000 (Ye et al. 2026)
- BEELINE mDC, TFs+500 (Ye et al. 2026)
- BEELINE mDC, TFs+1000 (Ye et al. 2026)
- BEELINE mHSC-E, TFs+500 (Ye et al. 2026)
- BEELINE mHSC-E, TFs+1000 (Ye et al. 2026)
- BEELINE mHSC-GM, TFs+500 (Ye et al. 2026)
- BEELINE mHSC-GM, TFs+1000 (Ye et al. 2026)
- BEELINE mHSC-L, TFs+500 (Ye et al. 2026)
- BEELINE mHSC-L, TFs+1000 (Ye et al. 2026)
- Six BEELINE datasets, TFs+500 (Ye et al. 2026)
- Six BEELINE datasets, TFs+1000 (Ye et al. 2026)
Evaluations (1,176)
- DeepSEM GRN on hESC, TFs+500, cell-type-specific ChIP-seq
- DeepSEM GRN on hESC, TFs+500, non-specific ChIP-seq
- DeepSEM GRN on hESC, TFs+500, OmniPath
- DeepSEM GRN on hESC, TFs+500, STRING
- DeepSEM GRN on hESC, TFs+1000, cell-type-specific ChIP-seq
- DeepSEM GRN on hESC, TFs+1000, non-specific ChIP-seq
- DeepSEM GRN on hESC, TFs+1000, OmniPath
- DeepSEM GRN on hESC, TFs+1000, STRING
- DeepSEM GRN on hHep, TFs+500, cell-type-specific ChIP-seq
- DeepSEM GRN on hHep, TFs+500, non-specific ChIP-seq
- DeepSEM GRN on hHep, TFs+500, OmniPath
- DeepSEM GRN on hHep, TFs+500, STRING
- DeepSEM GRN on hHep, TFs+1000, cell-type-specific ChIP-seq
- DeepSEM GRN on hHep, TFs+1000, non-specific ChIP-seq
- DeepSEM GRN on hHep, TFs+1000, OmniPath
- DeepSEM GRN on hHep, TFs+1000, STRING
- DeepSEM GRN on mDC, TFs+500, cell-type-specific ChIP-seq
- DeepSEM GRN on mDC, TFs+500, non-specific ChIP-seq
- DeepSEM GRN on mDC, TFs+500, OmniPath
- DeepSEM GRN on mDC, TFs+500, STRING
- DeepSEM GRN on mDC, TFs+1000, cell-type-specific ChIP-seq
- DeepSEM GRN on mDC, TFs+1000, non-specific ChIP-seq
- DeepSEM GRN on mDC, TFs+1000, OmniPath
- DeepSEM GRN on mDC, TFs+1000, STRING
- DeepSEM GRN on mHSC-E, TFs+500, cell-type-specific ChIP-seq
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.
25 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Artifact SHA-256 a6c3aae569b1f1614c71569efb0692f15700aea473c64c7bc0bc660e81330dc7 Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Artifact url https://www.biorxiv.org/content/10.64898/2026.05.17.725083v1.full.pdf Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Doi 10.64898/2026.05.17.725083 Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
Evidence concerns3 values
| Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Hash scope Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Licence CC-BY-4.0 Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
Limitations3 values
| Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Media type application/pdf Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Publication status preprint Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
| Retrieved at 2026-10-11T21:36:31Z Source metadata | Benchmarking Static Gene Regulatory Network Reconstruction and Dynamic Transition Probing in Single-Cell Foundation Models No field-specific location recorded Version: bioRxiv preprint version 1, posted 2026-05-20 | catalogued No individual claim review recorded Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: Source artifact SHA-256: Hash scope: Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference. |
Sources and history
Release 2026-10-23-5347695500ed · Record review: source checked
bioRxiv PDF, retrieved and hashed for the 2026-10-11 networks-mechanistic literature pass.
Technical metadata and extraction receipts
Stable ID: lit-cells-networks-20261011-ye2026grn-source-ye2026
- areas
- biological-networks; single-cell
- contexts
- research
- url
- https://doi.org/10.64898/2026.05.17.725083
- artifact url
- https://www.biorxiv.org/content/10.64898/2026.05.17.725083v1.full.pdf
- version
- bioRxiv preprint version 1, posted 2026-05-20
- retrieved at
- 2026-10-11T21:36:31Z
- artifact sha256
- a6c3aae569b1f1614c71569efb0692f15700aea473c64c7bc0bc660e81330dc7
- doi
- 10.64898/2026.05.17.725083
- publication status
- preprint
- licence
- CC-BY-4.0
- media type
- application/pdf
- hash scope
- Exact bytes of one download. bioRxiv serves slightly different PDF bytes per download (41 bytes differed between two downloads 2 minutes apart; pdftotext output identical), so the gzip copy in artifacts/ is the reference.
- limitations
- Preprint, not peer reviewed.; One run per dataset; the printed spread is the standard deviation across the six datasets, not across runs.; The authors developed none of the evaluated models or baselines.
- evidence concerns
- source id: lit-cells-networks-20261011-ye2026grn-source-ye2026; message: Table 1 labels its middle reference column 'ChIP-seq' and the caption names cell-type-specific ChIP-seq, but its values equal the 'Nonspecific ChIP-seq' mean and std rows of Tables 4 and 6 (for example scGPT costok EPR 4.211±1.703), and Appendix B says cell-type-specific ChIP-seq is excluded from the main comparison. Table 1 is therefore not extracted; Tables 4-7 carry the reference labels.; source locator: Table 1 caption and column header; Tables 4 and 6, 'Nonspecific ChIP-seq' mean rows; Appendix B text before Table 4; artifact sha256: a6c3aae569b1f1614c71569efb0692f15700aea473c64c7bc0bc660e81330dc7; review date: 2026-10-11; source id: lit-cells-networks-20261011-ye2026grn-source-ye2026; message: Table 6 is titled 'AUPR' and Table 7 'AUPRC', but both print ratios over a random predictor (values above 1), matching the 'AUPRC ratio' of Table 1; both are recorded as auprc-ratio.; source locator: Table 6 and Table 7 captions; Table 1 caption; artifact sha256: a6c3aae569b1f1614c71569efb0692f15700aea473c64c7bc0bc660e81330dc7; review date: 2026-10-11; source id: lit-cells-networks-20261011-ye2026grn-source-ye2026; message: Table 7 (TFs+1000 AUPRC) repeats Table 6 (TFs+500 AUPR) exactly in the GENIE3 column of the STRING, Cell-type-specific ChIP-seq and Nonspecific ChIP-seq blocks: all 18 dataset values and the three mean and std rows are identical (for example STRING hESC 1.6241, mean 3.301, std 2.244). Only the OmniPath block differs. Every other column, and GENIE3 in the EPR Tables 4 and 5, differs between the two gene sets, and the Table 6 values match Table 1. These 21 Table 7 GENIE3 values are likely a copy of the TFs+500 values and should not be compared.; source locator: Table 7, column group 'ML', column 'GENIE3', blocks 'STRING', 'Cell-type-specific' and 'Nonspecific'; Table 6, same cells; Table 1, GENIE3 AUPRC ratio; artifact sha256: a6c3aae569b1f1614c71569efb0692f15700aea473c64c7bc0bc660e81330dc7; review date: 2026-10-12
- review
- method: source-hash-verification; method note: Archived gzip copy unpacked and its SHA-256 matched; a fresh bioRxiv download was rate-limited (HTTP 429).; reviewer: claude; reviewer note: Claude review agent reviewer-cn-1, independent of the extractor lit-cells-networks; no human review claimed; date: 2026-10-12