Model

ProkBERT

ProkBERT is a family of microbial DNA encoders used for sequence representation, promoter prediction and phage identification.

Sources (4)

nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata

3 evaluations · 12 metric rows · 3 evaluated configurations using this model

How it worksProkBERT workflow

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

Sources (4)

Overview

Model type

Microbial DNA BERT encoder

Sources (4)

Inputs

Microbial DNA segments tokenized with the selected variant.

Sources (4)

Outputs

Sequence representations or predictions from separately fine-tuned promoter/phage heads.

Sources (4)

Access

Official project documentation and implementation: https://github.com/nbrg-ppcu/prokbert

Sources (4)

Source reviewed · Automated source review, 2026-09-16. All specifications and missing details

Evaluations and results

3 evaluations · 12 metric rows. Different protocols are not a single leaderboard.

Filter evaluations

Tested model or methodProtocolDatasetEvaluation setupMetricEvidence origin

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configuration	Protocol and dataset	Finding	Evidence and details
Configuration: ProkBERT-mini	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.87 Accuracy unitless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini: E. coli sigma70 promoter prediction Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications; ProkBERT family: genomic language models for microbiome applications · Table 3, ProkBERT-mini row, Accuracy column
Configuration: ProkBERT-mini-long	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.89 Sensitivity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-long: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-long, column Sensitivity; XML row4 column4
Configuration: ProkBERT-mini	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.90 Sensitivity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini: E. coli sigma70 promoter prediction Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini, column Sensitivity; XML row2 column4
Configuration: ProkBERT-mini	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.85 Specificity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini: E. coli sigma70 promoter prediction Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini, column Specificity; XML row2 column5
Configuration: ProkBERT-mini-c	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.88 Sensitivity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-c: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-c, column Sensitivity; XML row3 column4
Configuration: ProkBERT-mini-c	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.85 Specificity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-c: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-c, column Specificity; XML row3 column5
Configuration: ProkBERT-mini-long	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.87 Accuracy unitless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-long: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-long, column Accuracy; XML row4 column2
Configuration: ProkBERT-mini-c	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.73 MCC dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-c: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-c, column MCC; XML row3 column3
Configuration: ProkBERT-mini-long	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.85 Specificity fraction · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-long: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-long, column Specificity; XML row4 column5
Configuration: ProkBERT-mini-long	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.74 MCC dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-long: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-long, column MCC; XML row4 column3
Configuration: ProkBERT-mini-c	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.87 Accuracy unitless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini-c: E. coli sigma70 independent promoter test Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini-c, column Accuracy; XML row3 column2
Configuration: ProkBERT-mini	Protocol: E. coli sigma70 independent promoter test (E. coli sigma70 promoter prediction) Dataset: E. coli sigma70 promoter dataset	0.74 MCC dimensionless · higher Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible	Author-reported evaluation · source checked Methods, coverage and source ProkBERT-mini: E. coli sigma70 promoter prediction Test-only evaluation on Cassiano and Silva-Rocha 2020 data; methods have different training histories. 865 high-evidence RegulonDB 10.5 promoters and 1,000 nucleotide-distribution-matched negative sequences. Promoter exact matches removed from model training. Aggregation: Not reported ProkBERT family: genomic language models for microbiome applications · Table 3, row ProkBERT-mini, column MCC; XML row2 column3

Source checking is not independent reproduction. Release 2026-09-23-2b89723c6dd9.

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.

ProkBERT-mini-long† Pipeline · 6 metric rows
ProkBERT-mini-c† Pipeline · 6 metric rows
ProkBERT-mini† Pipeline · 6 metric rows

Use this model

How it works, versions and access

Versions and evaluated configurations

ProkBERT-mini-c · family
ProkBERT-mini-long · family
ProkBERT-mini · family

How it works

ProkBERT is a family of microbial DNA encoders used for sequence representation, promoter prediction and phage identification. MegatronBERT-based encoder using local-context-aware k-mer tokenization, learned relative key/value positions and masked-language pretraining. The documented inputs are microbial DNA segments tokenized with the selected variant. The output consists of sequence representations or predictions from separately fine-tuned promoter/phage heads.

Sources (4)

Versions and reproducibility

ProkBERT-mini (6-mer, shift 1), mini-c (single base), mini-long (6-mer, shift 2), plus promoter/phage fine-tunes. The paper reports maximum sequence lengths of 1,024bp for mini and 2,048bp for mini-long; tokenization stride distinguishes the variants.

Sources (4)

Strengths, limitations and unresolved questions

Strengths and limitations

Strengths supported by sources

The released family includes character and shifted k-mer variants plus task-specific fine-tuned models.
Sources (4)
nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata

Limitations and conditions

The authors explicitly identify restricted context as a limitation in phage analysis. A promoter or phage classifier is a distinct fine-tuned configuration.
Sources (4)
nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata

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-prokbert

Specifications

Inputs, training, access and other details

Explanatory profile: source reviewed · 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
Property	Description and evidence
Model type	Microbial DNA BERT encoder Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Architecture	MegatronBERT-based encoder using local-context-aware k-mer tokenization, learned relative key/value positions and masked-language pretraining. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Inputs	Microbial DNA segments tokenized with the selected variant. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Outputs	Sequence representations or predictions from separately fine-tuned promoter/phage heads. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Parameters	20.6M for the inspected ProkBERT-mini checkpoint; do not assign that count automatically to all variants. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Known versions	ProkBERT-mini (6-mer, shift 1), mini-c (single base), mini-long (6-mer, shift 2), plus promoter/phage fine-tunes. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Training data	NCBI RefSeq genomes covering bacteria, viruses, archaea and fungi; README reports 976,878 contigs from 17,178 assemblies and 3,882 genera. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Training cutoff	NCBI RefSeq training-genome retrieval date: 6 January 2023, as reported in Methods. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Context limits	The paper reports maximum sequence lengths of 1,024bp for mini and 2,048bp for mini-long; tokenization stride distinguishes the variants. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Weights licence	The official neuralbioinfo/prokbert-mini card declares CC-BY-NC-4.0. The code repository is MIT, so code and model-weight rights differ. Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Access	Official project documentation and implementation: https://github.com/nbrg-ppcu/prokbert Sources (4) nbrg-ppcu/prokbert: README.md; neuralbioinfo/prokbert-mini: README.md; neuralbioinfo/prokbert-mini: config.json; prokbert: Journal full-text XML · ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata
Code licence	MIT Sources nbrg-ppcu/prokbert: 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.

77 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-23-2b89723c6dd9
Property and statement	Original source and location	Review 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	nbrg-ppcu/prokbert: README.md Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 8670ae92b816cff158a0b85647a8dea122e251eb Retrieved: 2026-09-16T19:46:19.913949+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: `29de39c6ad006ce704ab14240cfd97af93da411fb63ec89ebe499c2646928cfc` 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	prokbert: Journal full-text XML Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied Retrieved: 2026-09-16T19:53:03.847832+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: `8610e2a54aa877c8dc565a9cdb6e82099f284c5e0907a52cab18d994ea732436` 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	neuralbioinfo/prokbert-mini: README.md Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc Retrieved: 2026-09-16T20:04:02.231162+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: `172ec5b600e342302df5a22d84b12e1ab61d05722adda549b6d82f65e5c5d658` 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	neuralbioinfo/prokbert-mini: config.json Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc Retrieved: 2026-09-16T20:04:02.231162+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: `6d68c922f919e0de3f87de60c07ebc1823927df806645937f94911c5b8245de6` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram steps DNA segment Variant-specific LCA tokens BERT encoder Representation or task head Individual claims	nbrg-ppcu/prokbert: README.md Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 8670ae92b816cff158a0b85647a8dea122e251eb Retrieved: 2026-09-16T19:46:19.913949+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: `29de39c6ad006ce704ab14240cfd97af93da411fb63ec89ebe499c2646928cfc` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram steps DNA segment Variant-specific LCA tokens BERT encoder Representation or task head Individual claims	prokbert: Journal full-text XML Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied Retrieved: 2026-09-16T19:53:03.847832+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: `8610e2a54aa877c8dc565a9cdb6e82099f284c5e0907a52cab18d994ea732436` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram steps DNA segment Variant-specific LCA tokens BERT encoder Representation or task head Individual claims	neuralbioinfo/prokbert-mini: README.md Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc Retrieved: 2026-09-16T20:04:02.231162+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: `172ec5b600e342302df5a22d84b12e1ab61d05722adda549b6d82f65e5c5d658` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram steps DNA segment Variant-specific LCA tokens BERT encoder Representation or task head Individual claims	neuralbioinfo/prokbert-mini: config.json Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc Retrieved: 2026-09-16T20:04:02.231162+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: `6d68c922f919e0de3f87de60c07ebc1823927df806645937f94911c5b8245de6` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram title ProkBERT workflow Individual claims	nbrg-ppcu/prokbert: README.md Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: 8670ae92b816cff158a0b85647a8dea122e251eb Retrieved: 2026-09-16T19:46:19.913949+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: `29de39c6ad006ce704ab14240cfd97af93da411fb63ec89ebe499c2646928cfc` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact
Diagram title ProkBERT workflow Individual claims	prokbert: Journal full-text XML Original source ↗ ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata Shared locator for this statement’s cited sources; not a separate locator for each citation. Version: Retrieved page snapshot; no immutable publisher revision supplied Retrieved: 2026-09-16T19:53:03.847832+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: `8610e2a54aa877c8dc565a9cdb6e82099f284c5e0907a52cab18d994ea732436` Hash scope: SHA-256 of retrieved original artifact bytes Format: original_artifact Inspected artifact

All evidence (CSV)All evidence (JSONL)How to read the evidence

Sources and history

View linked audit checks and correction history

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

6 source records and release history

ProkBERT official resource · Original source · Not reported
nbrg-ppcu/prokbert: README.md · Original source · 8670ae92b816cff158a0b85647a8dea122e251eb
neuralbioinfo/prokbert-mini: README.md · Original source · feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc
neuralbioinfo/prokbert-mini: config.json · Original source · feb2520a43cd9cdb5b3d8477e47209dbcb55d1dc
prokbert: Journal full-text XML · Original source · Retrieved page snapshot; no immutable publisher revision supplied
nbrg-ppcu/prokbert: LICENSE · Original source · 8670ae92b816cff158a0b85647a8dea122e251eb

Download this release

Technical metadata and extraction receipts

Stable ID: catalog-model-prokbert

areas: microbes-communities
method types: foundation model
entity level: family
version: mini
reported name: ProkBERT
access: Public model family and mini checkpoint.
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-2efbfaba5a1c09f4f7fa; evidence-official-965600234c25fbd104c8; evidence-official-ee6e270af125ddf16eb3; evidence-official-5e268f31f0347fc5564c; source locator: ProkBERT paper Section 2.1.2 and Table 1; official ProkBERT-mini card: Model Description and licence metadata; ambiguities: None recorded

Related records

Suggest a correction