Datasets
Curated phage and bacterial reference-genome collections.
Task
Genome-wide prophage detection
Prophage detection is evaluated with grouped data partitions and explicit sequence-composition controls.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passagesOverview
Metrics
Separate control collections assess bacterial false positives and phage false negatives.
Allowed inputs
Genomic sequence.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passagesEvaluation procedure diagram
How it worksComputational evaluation flow
Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passageslimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Results
Each comparison retains its reviewed evaluation scope, dataset and metric. Results are shown without a pooled ranking.
Genome-wide prophage detection · Table 5.
Precision (fraction) · Higher values are better.
Genome-wide prophage detection (Genome-wide prophage detection) · LAMBDA genome-wide prophage test
Evidence origin: Independent external evaluation, Author-reported evaluation.
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Table 5.: Precision, Genome-wide prophage detection- Different tool input pipelines and training histories; †gLM and‡pLM markers retained.
- Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.
Comparison details and limitations
Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.
- No interval assigned unless printed in source cell.
Automated source review: 2026-09-17. Numerical source review does not establish independent reproduction.
Dots show point estimates. Whiskers show only explicitly defined uncertainty (standard deviation, standard error or a labelled interval); their definitions remain in Table. Unresolved uncertainty is not plotted. Differences do not establish statistical significance.
Showing 12 of 17 matching rows.
Methods and evaluation design
Procedure, tasks and evaluated configurations
How it works
Evaluation methodology
Curated phage and bacterial reference-genome collections. Cluster/group-aware 80:10:10 training, development and test split. Separate control collections assess bacterial false positives and phage false negatives. Phage clusters and bacterial genus groups are assigned wholly to one partition. Repeated experiments are summarized with mean and standard deviation for each compared model.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passagesEvaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
These source-backed links do not make different protocols or scores interchangeable.
Run this benchmark
Choose a concrete protocol before running an evaluation. Its inputs, split and scoring rules determine which results can be compared.
Run instructions
No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.
Strengths, limitations and unresolved questions
Strengths and limitations
Strengths and considerations
- Cluster/group-aware partitioning addresses related-sequence overlap between training and testing.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages
Limitations and conditions
- Shuffled-sequence controls test sequence-order dependence while preserving composition. Their classification scores and genome-wide localization comparisons are separate endpoints.
Sources
LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages
Profile review details
Relevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged.
Stable record: reported-task-dd001540e0f4ecSpecifications
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.
| Property | Description and evidence |
|---|---|
| Datasets | Curated phage and bacterial reference-genome collections.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Splits | Cluster/group-aware 80:10:10 training, development and test split.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Metrics | Separate control collections assess bacterial false positives and phage false negatives.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Baselines | Genome-wide comparisons include PHASTER, geNomad, VIBRANT, Phigaro, PhiSpy and VirSorter2, plus protein-language-model-based PIDE. Embedding probes also compare pretrained and randomly initialized representations; those probes are a separate comparison from genome-wide tools.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Comparison with Traditional and Protein-based Models; embedding-probe experiment and Tables 2–3 |
| Leakage controls | Phage clusters and bacterial genus groups are assigned wholly to one partition.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Uncertainty | Repeated experiments are summarized with mean and standard deviation for each compared model.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Entity type | Paper-specific computational evaluation protocol.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Organisms | Phage and bacterial reference collections.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Assays | Reference genome/prophage annotations.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Allowed inputs | Genomic sequence.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
| Adaptation | Supervised detection with a cluster/group-aware train/development/test split.SourcesLAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages |
Evidence
Source checking verifies the cited claim or transcription. It does not establish independent reproduction.
Papers and result coverage
Last literature check: 2026-09-17. Primary-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.
| Paper or primary resource | Version | Reference |
|---|---|---|
| LAMBDA: A Prophage Detection Benchmark for Genomic Language Models | PMC13041943.1 | Read source DOI: 10.64898/2026.03.26.714501 |
Historical gaps recorded on 2026-09-17
The catalogue now holds 102 result rows for this benchmark. A note below about pending extraction describes the state on 2026-09-17 and may since have been answered by a later batch. The result rows and their sources are the current record.
- Independent batch review before import; preserve existing observation identities.
Search and extraction details
complete comparison tables extracted pending publication review
Searches
- LAMBDA: A Prophage Detection Benchmark for Genomic Language Models primary paper benchmark results
Evidence locations
- Table5 genome-wide versus Table3fragmentexperiment; reference construction
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.
18 evidence rows matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Diagram caption Conceptual summary of the cited evaluation; exact task configuration and source version remain part of the protocol. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Diagram steps
| LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Computational evaluation flow Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Datasets Curated phage and bacterial reference-genome collections. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Splits Cluster/group-aware 80:10:10 training, development and test split. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Adaptation Supervised detection with a cluster/group-aware train/development/test split. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Metrics Separate control collections assess bacterial false positives and phage false negatives. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Baselines Genome-wide comparisons include PHASTER, geNomad, VIBRANT, Phigaro, PhiSpy and VirSorter2, plus protein-language-model-based PIDE. Embedding probes also compare pretrained and randomly initialized representations; those probes are a separate comparison from genome-wide tools. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Comparison with Traditional and Protein-based Models; embedding-probe experiment and Tables 2–3 Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Leakage controls Phage clusters and bacterial genus groups are assigned wholly to one partition. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Uncertainty Repeated experiments are summarized with mean and standard deviation for each compared model. Individual claims | LAMBDA: A Prophage Detection Benchmark for Genomic Language Models Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages Version: PMC13041943.1 | source checked automated source review · 2026-09-16 Audit detailsRelevant full-paper computational evaluation sections, tables/captions and cited supplementary task passages were reviewed. Reporting omissions are scoped to the inspected sources. Original numerical results are unchanged. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Sources and history
View linked audit checks and correction history
Release 2026-09-29-06401fd5b220 · Record review: needs review
2 source records and release history
- LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Original source · PMC13041943.1
- LAMBDA: A Prophage Detection Benchmark for Genomic Language Models · Original source · PMC13041943.1
Technical metadata and extraction receipts
Stable ID: reported-task-dd001540e0f4ec
- areas
- microbes-communities
- tasks
- Genome-wide prophage detection
- entity level
- task
- version
- Not reported
- task
- Genome-wide prophage detection
- scope note
- Paper-specific evaluation task; protocol completeness requires further extraction.
- comparison panels
- id: part2-lambda-prophage-2026-T5-ea2b9e053a; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: Precision; unit: fraction; direction: higher; result ids: paper-result-e7a3a4d64ab390ab49; paper-result-a06603949a50933a8d; paper-result-c802c8c15e3405562c; paper-result-74849712f209d7ac30; paper-result-8e6b2821ac9808600a; paper-result-79a061473b6c519bbc; paper-result-9f8836b9f95e5299cd; paper-result-c1547589ceea1d9e54; paper-result-31813172c31b723177; paper-result-d539bf7c128bc6836a; paper-result-dad89fd9d2431eb25f; paper-result-911d0027540ad9a5d5; paper-result-181b0c4d8249ebadfe; paper-result-08e57fd4a4dfc1d04a; paper-result-85552cecec875a2a2d; paper-result-be6edcd8da058a2a18; paper-result-2532be2c98577da641; source ids: part2-lambda-prophage-2026; source locator: Table 5.: Precision, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-lambda-prophage-2026-T5-53308364e0; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: Recall; unit: fraction; direction: higher; result ids: paper-result-97e52289280e1fae96; paper-result-cdb7fa6252004e69a0; paper-result-d2671f470add052c95; paper-result-b055c366896ed5de0f; paper-result-a6a13a958788577c38; paper-result-f115eb38187794e35f; paper-result-44f9400644e72d0d06; paper-result-b1d0506bc82f67b1eb; paper-result-497e6cae9ab0ad3020; paper-result-296e3f408bd35593be; paper-result-df8cde861fc3b93615; paper-result-8ae7ec3f68a41210d2; paper-result-62f561dad5ccb58842; paper-result-6da890addbc1047a27; paper-result-40524c22a54dc0a9ea; paper-result-86d4ac1a8cf79fcb61; paper-result-b507c4759c12caf2ff; source ids: part2-lambda-prophage-2026; source locator: Table 5.: Recall, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-lambda-prophage-2026-T5-5384a9619c; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: Specificity; unit: fraction; direction: higher; result ids: paper-result-e14d406e39fc25afe5; paper-result-b6f96d58c259c3ddbc; paper-result-c0332beda4d25cf031; paper-result-998cbe6bfc62cf0dc2; paper-result-11b156230bc9c72f52; paper-result-e529fb96fa1afd802f; paper-result-e8ad7325a802539ff7; paper-result-4fca80021785b3b17d; paper-result-9759b1c8ba49d0637c; paper-result-ad193cda382dc5f155; paper-result-c9c3c03da0764cd397; paper-result-dfc1dd7998a8f0b87f; paper-result-c42e3020111d6d6d92; paper-result-bfb95eb33658a02228; paper-result-9f218b36b330581b4e; paper-result-0db65312a74302c448; paper-result-edcf8c457f3b726d0c; source ids: part2-lambda-prophage-2026; source locator: Table 5.: Specificity, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-lambda-prophage-2026-T5-906d46edf7; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: FPR; unit: fraction; direction: lower; result ids: paper-result-64d71695f6ecff7d37; paper-result-394b7604923b5adc9e; paper-result-4817f5ddcd632a2f5a; paper-result-f28d464414ab011128; paper-result-dcef5aaee3067608f0; paper-result-4f8c584360869901f5; paper-result-87215da087c0b15cfa; paper-result-d8f68c1b918eed398d; paper-result-7d20b1be5cc6e8043a; paper-result-33209fc0bc77878645; paper-result-f46c39cbbb69ebaf6d; paper-result-dc81699af30d77270d; paper-result-36e647332f530a2641; paper-result-114205a0aa30717ddd; paper-result-23456ae224ed60c7f6; paper-result-1a8de38949de72ecf2; paper-result-0261b022358700df1f; source ids: part2-lambda-prophage-2026; source locator: Table 5.: FPR, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-lambda-prophage-2026-T5-1f80d97aab; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: F1; unit: fraction; direction: higher; result ids: paper-result-2c8d80c884705f8507; paper-result-efe194ad9f2e77e0cd; paper-result-cbd98a692f3688820b; paper-result-80b1d1f3626c4fdd21; paper-result-66bb9bf85666106228; paper-result-d42a3d827c6a3e8fbf; paper-result-d06ddd98d6567adf47; paper-result-b19996145b265cbe13; paper-result-0f2fe445b92f4f28de; paper-result-06e7cc4727bad189fe; paper-result-1bb7a28861bb8e5821; paper-result-aeb146019fa85bcdf2; paper-result-b133e6df703f7611a7; paper-result-5db8cdaca7588fa5ca; paper-result-6ce4c22c1bee5109ba; paper-result-81f827e7549c692370; paper-result-84ce8d28189999f1bd; source ids: part2-lambda-prophage-2026; source locator: Table 5.: F1, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17; id: part2-lambda-prophage-2026-T5-27d631c420; title: Genome-wide prophage detection · Table 5.; protocol id: paper-protocol-41c6e227215346177d; dataset id: reported-dataset-1b4f6ea24c0587; metric: MCC; unit: unitless; direction: higher; result ids: lit-038; paper-result-ad8f7a939fe6fedda9; paper-result-a65058240bc1b8e345; paper-result-4615d60b969151525b; paper-result-1fb211666087dddc72; lit-037; paper-result-02e0fc1798b036d3d3; paper-result-ffa067657fc300a3b6; paper-result-44767fd81dbbcd9a15; paper-result-b7a2e1cb2f2c95a963; paper-result-5f92e457e48dca6143; paper-result-6104a0c48f501a442a; paper-result-dcde4b7b93eec8e1ce; paper-result-6ac04b40baae819886; paper-result-dae3fa85e4aea76696; paper-result-9e2cfcc5dcc55abc03; paper-result-b903ac601cf415526f; source ids: part2-lambda-prophage-2026; source locator: Table 5.: MCC, Genome-wide prophage detection; context: Scan complete genomes with overlapping windows and postprocess predictions into prophage regions. Genome-wide metrics are computed at region level and macro-averaged across genomes. Filtered gLM pipelines include normalization, smoothing, clustering and size filtering; these are not standalone encoder scores. 80 bacterial genomes,47 species,four phyla;386 annotated prophage locations.; caveats: Different tool input pipelines and training histories; †gLM and‡pLM markers retained.; Table5 reports point estimates; ten repeats belong to separate Table3 fragment experiment.; No interval assigned unless printed in source cell.; review: method: automated_source_review; date: 2026-09-17
- benchmark research
- review date: 2026-09-17; status: complete_comparison_tables_extracted_pending_publication_review; primary sources: part2-lambda-prophage-2026; inspected locators: Table5 genome-wide versus Table3fragmentexperiment; reference construction; searched queries: LAMBDA: A Prophage Detection Benchmark for Genomic Language Models primary paper benchmark results; gaps: Independent batch review before import; preserve existing observation identities.; claim scope: Primary-source discovery and table/protocol screening; source checked is not independently reproduced. Raw acquisitions not automatically numerical publication approval.
- historical missing metadata
- protocol version: not_reported_in_legacy_extract; split: not_reported_in_legacy_extract
- 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.
- legacy kinds
- benchmark
- entity classification
- review date: 2026-09-17; rationale: This source-scoped record identifies the biological prediction task and holds its paper context. Preserve the existing task identity; exact split, model adaptation and scoring remain in linked evaluations or separate protocol records.; source ids: lambda-prophage-2026; source locator: Methods: LAMBDA Benchmark Dataset Construction; Data Splitting and Data Leakage Prevention; control datasets; cached text lines 66–67, 78–85; uncertainty/repeat-run/statistical-comparison passages; ambiguities: A paper- or suite-specific task may constrain some inputs or metrics; that alone does not make it interchangeable with a complete versioned protocol. No protocol equivalence is inferred.; Some legacy profile Entity type facts use the generic phrase computational evaluation protocol. That boilerplate is not sufficient to establish a single fixed protocol identity or to merge this task with another protocol record.
Related records
- dataset: LAMBDA genome-wide prophage test
- evaluates task: Genome-wide prophage detection (Genome-wide prophage detection)