rewirebio.iobenchmarks
← Browse all sources
Source

SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Nature Communications · 2026 · Published article, Nature Communications 2026 (PMC13342640.1) · Peer-reviewed

DOI: 10.1038/s41467-026-72570-x

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.

Model or configurationMetricValueDatasetEvaluation
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.6603 unitless10x-Multiome-Pbmc10k (SCMBench)bindSC on SCMBench paired integration, 10x-Multiome-Pbmc10k
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.352475 unitlessChen-2019 (SCMBench)bindSC on SCMBench paired integration, Chen-2019
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.5579 unitlessMa-2020-batch-53 (SCMBench)bindSC on SCMBench paired integration, Ma-2020-batch-53
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.474975 unitlessMa-2020-batch-54 (SCMBench)bindSC on SCMBench paired integration, Ma-2020-batch-54
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.4826 unitlessMa-2020-batch-55 (SCMBench)bindSC on SCMBench paired integration, Ma-2020-batch-55
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.50565 unitlesspaired-overall (SCMBench)bindSC on SCMBench paired integration, paired-overall
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.6708 unitlessMuto-2021-batch-1 (SCMBench)bindSC on SCMBench unpaired integration, Muto-2021-batch-1
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.7157 unitlessMuto-2021-batch-2 (SCMBench)bindSC on SCMBench unpaired integration, Muto-2021-batch-2
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.711375 unitlessMuto-2021-batch-3 (SCMBench)bindSC on SCMBench unpaired integration, Muto-2021-batch-3
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.76825 unitlessMuto-2021-batch-4 (SCMBench)bindSC on SCMBench unpaired integration, Muto-2021-batch-4
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.68975 unitlessMuto-2021-batch-5 (SCMBench)bindSC on SCMBench unpaired integration, Muto-2021-batch-5
bindSC (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across the five Muto-2021 batches (Yao not scored))0.711175 unitlessunpaired-overall (SCMBench)bindSC on SCMBench unpaired integration, unpaired-overall
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.630875 unitless10x-Multiome-Pbmc10k (SCMBench)Cobolt on SCMBench paired integration, 10x-Multiome-Pbmc10k
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.546175 unitlessChen-2019 (SCMBench)Cobolt on SCMBench paired integration, Chen-2019
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.495775 unitlessMa-2020-batch-53 (SCMBench)Cobolt on SCMBench paired integration, Ma-2020-batch-53
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.496125 unitlessMa-2020-batch-54 (SCMBench)Cobolt on SCMBench paired integration, Ma-2020-batch-54
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.495475 unitlessMa-2020-batch-55 (SCMBench)Cobolt on SCMBench paired integration, Ma-2020-batch-55
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.532885 unitlesspaired-overall (SCMBench)Cobolt on SCMBench paired integration, paired-overall
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.314775 unitlessMuto-2021-batch-1 (SCMBench)Cobolt on SCMBench unpaired integration, Muto-2021-batch-1
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.372825 unitlessMuto-2021-batch-2 (SCMBench)Cobolt on SCMBench unpaired integration, Muto-2021-batch-2
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.2894 unitlessMuto-2021-batch-3 (SCMBench)Cobolt on SCMBench unpaired integration, Muto-2021-batch-3
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.408275 unitlessMuto-2021-batch-4 (SCMBench)Cobolt on SCMBench unpaired integration, Muto-2021-batch-4
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.315825 unitlessMuto-2021-batch-5 (SCMBench)Cobolt on SCMBench unpaired integration, Muto-2021-batch-5
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.3692375 unitlessunpaired-overall (SCMBench)Cobolt on SCMBench unpaired integration, unpaired-overall
Cobolt (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.514325 unitlessYao (SCMBench)Cobolt on SCMBench unpaired integration, Yao
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.56745 unitless10x-Multiome-Pbmc10k (SCMBench)Deepmaps on SCMBench paired integration, 10x-Multiome-Pbmc10k
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.48425 unitlessChen-2019 (SCMBench)Deepmaps on SCMBench paired integration, Chen-2019
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.272475 unitlessMa-2020-batch-53 (SCMBench)Deepmaps on SCMBench paired integration, Ma-2020-batch-53
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.232325 unitlessMa-2020-batch-54 (SCMBench)Deepmaps on SCMBench paired integration, Ma-2020-batch-54
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.216375 unitlessMa-2020-batch-55 (SCMBench)Deepmaps on SCMBench paired integration, Ma-2020-batch-55
Deepmaps (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.354575 unitlesspaired-overall (SCMBench)Deepmaps on SCMBench paired integration, paired-overall
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.673625 unitless10x-Multiome-Pbmc10k (SCMBench)FM-scVI on SCMBench paired integration, 10x-Multiome-Pbmc10k
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.56805 unitlessChen-2019 (SCMBench)FM-scVI on SCMBench paired integration, Chen-2019
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.278275 unitlessMa-2020-batch-53 (SCMBench)FM-scVI on SCMBench paired integration, Ma-2020-batch-53
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.2949 unitlessMa-2020-batch-54 (SCMBench)FM-scVI on SCMBench paired integration, Ma-2020-batch-54
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.29955 unitlessMa-2020-batch-55 (SCMBench)FM-scVI on SCMBench paired integration, Ma-2020-batch-55
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.42288 unitlesspaired-overall (SCMBench)FM-scVI on SCMBench paired integration, paired-overall
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.57645 unitlessMuto-2021-batch-1 (SCMBench)FM-scVI on SCMBench unpaired integration, Muto-2021-batch-1
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.5988 unitlessMuto-2021-batch-2 (SCMBench)FM-scVI on SCMBench unpaired integration, Muto-2021-batch-2
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.6026 unitlessMuto-2021-batch-3 (SCMBench)FM-scVI on SCMBench unpaired integration, Muto-2021-batch-3
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.580025 unitlessMuto-2021-batch-4 (SCMBench)FM-scVI on SCMBench unpaired integration, Muto-2021-batch-4
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.550475 unitlessMuto-2021-batch-5 (SCMBench)FM-scVI on SCMBench unpaired integration, Muto-2021-batch-5
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across datasets)0.569729166666667 unitlessunpaired-overall (SCMBench)FM-scVI on SCMBench unpaired integration, unpaired-overall
FM-scVI (SCMBench adapter)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.510025 unitlessYao (SCMBench)FM-scVI on SCMBench unpaired integration, Yao
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.3564 unitlessMuto-2021-batch-1 (SCMBench)Geneformer on SCMBench unpaired integration, Muto-2021-batch-1
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.3817 unitlessMuto-2021-batch-2 (SCMBench)Geneformer on SCMBench unpaired integration, Muto-2021-batch-2
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.347925 unitlessMuto-2021-batch-3 (SCMBench)Geneformer on SCMBench unpaired integration, Muto-2021-batch-3
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.36815 unitlessMuto-2021-batch-4 (SCMBench)Geneformer on SCMBench unpaired integration, Muto-2021-batch-4
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI)0.335775 unitlessMuto-2021-batch-5 (SCMBench)Geneformer on SCMBench unpaired integration, Muto-2021-batch-5
Geneformer (SCMBench)Scmbench integration accuracy (mean of MAP, NMI, ASW and ARI; mean across the five Muto-2021 batches (Yao not scored))0.35799 unitlessunpaired-overall (SCMBench)Geneformer on SCMBench unpaired integration, unpaired-overall

Models, benchmarks and other records

Configurations (25)

Benchmarks (1)

Protocols (2)

Dataset subsets (13)

Evaluations (241)

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

Claims, original sources and review scope · Release 2026-10-23-5347695500ed
Property and statementOriginal source and locationReview and provenance
Artifact SHA-256
7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.artifact_sha256

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Artifact url
https://www.ebi.ac.uk/europepmc/webservices/rest/PMC13342640/fullTextXML
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.artifact_url

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Doi
10.1038/s41467-026-72570-x
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.doi

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Licence
CC-BY-4.0
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.licence

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Limitations
1 values
  • The authors developed the FM-scVI adapter strategy; all other methods are third-party.
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.limitations

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Media type
application/xml
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.media_type

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Publication status
peer_reviewed
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.publication_status

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Retrieved at
2026-10-11T21:38:51Z
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.retrieved_at

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Review: date
2026-10-12
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.review.date

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Review: method note
Artifact re-downloaded and SHA-256 matched; the archived gzip copy matches.
Source metadata		SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration

Original source ↗

No field-specific location recorded

Version: Published article, Nature Communications 2026 (PMC13342640.1)
Retrieved: 2026-10-11T21:38:51Z

catalogued

No individual claim review recorded

Audit details

Field: attributes.review.method_note

Source artifact SHA-256: 7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

How to read the evidence

Sources and history

Release 2026-10-23-5347695500ed · Record review: source checked

Primary source retrieved and hashed for the 2026-10-11 cells-spatial and networks-mechanistic literature pass.

Release history

No supporting source is linked yet.

Read original source

Download this release (gzip)
Technical metadata and extraction receipts

Stable ID: lit-cells-networks-20261011-scmbench-source-article

areas
single-cell
contexts
research
doi
10.1038/s41467-026-72570-x
publication status
peer_reviewed
licence
CC-BY-4.0
retrieved at
2026-10-11T21:38:51Z
year
2026
venue
Nature Communications
url
https://doi.org/10.1038/s41467-026-72570-x
artifact url
https://www.ebi.ac.uk/europepmc/webservices/rest/PMC13342640/fullTextXML
version
Published article, Nature Communications 2026 (PMC13342640.1)
artifact sha256
7aa2dbf1fc26bd602bf26e52c1061f91358a1c6cb389f57edc0ebfb0edc64f57
media type
application/xml
limitations
The authors developed the FM-scVI adapter strategy; all other methods are third-party.
review
method: source-hash-verification; method note: Artifact re-downloaded and SHA-256 matched; the archived gzip copy matches.; reviewer: claude; reviewer note: Claude review agent reviewer-cn-2, independent of the extractor lit-cells-networks; no human review claimed; date: 2026-10-12
Related records

    Suggest a correction