Datasets
HumanPPI downstream task; exact dataset release remains unextracted.
Task
human protein-protein interaction prediction
Human protein-interaction prediction evaluates structural augmentation of pretrained protein representations.
Sources
MULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passagesOverview
Metrics
Area under the ROC curve for binary protein-interaction classification.
Allowed inputs
Protein pairs represented using sequence and structure-informed embeddings.
Sources
MULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; 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
MULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passageslimited source coverage · Automated source review, 2026-09-16. All specifications and missing details
Results
Results are available, but no reviewed comparison panel is linked in this release.
All evaluations
1 evaluation · 1 result. Different protocols are not a single leaderboard.
Filter evaluations
Applied filters: All linked evaluations
| Tested configuration | Protocol and dataset | Finding | Evidence and details |
|---|---|---|---|
| Configuration: MULAN-ESM2 S | Task: human protein-protein interaction prediction Dataset: HumanPPI | 0.717 AUC fraction · unknown Uncertainty: Not reported Coverage: Not reported scored / Not reported eligible | Author-reported evaluation · Source checkedMethods, coverage and sourceMULAN-ESM2 S: human protein-protein interaction prediction MULAN sequence-structure model based on ESM2 8M Aggregation: Not reported MULAN: multimodal protein language model for sequence and structure encoding · Table 2, MULAN-ESM2 S row, HumanPPI AUC column |
Source checking is not independent reproduction. Release 2026-09-29-06401fd5b220.
Methods and evaluation design
Procedure, tasks and evaluated configurations
How it works
Evaluation methodology
HumanPPI downstream task; exact dataset release remains unextracted. Downstream hyperparameters and checkpoint selection use validation metrics. Area under the ROC curve for binary protein-interaction classification. MULAN-augmented models are compared with their underlying ESM2 representations. Original downstream dataset splits are retained; this alone does not certify protein-pair identity independence or absence from backbone pretraining. Downstream results average five initialization seeds; corresponding standard deviations are provided in the supplement.
Sources
MULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passagesRecorded evaluations
Each evaluation records what was tested and under which conditions.
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
No source-reviewed explanatory claims are recorded here yet.
Limitations and conditions
- Protein-pair grouping, negative sampling, metric implementation remain unextracted. Gains in other downstream tasks do not specify this task’s protocol.
Sources
MULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages
Profile review details
Task-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced.
Stable record: reported-task-6e54c7452b2b81Specifications
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 | HumanPPI downstream task; exact dataset release remains unextracted.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Splits | Downstream hyperparameters and checkpoint selection use validation metrics.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Metrics | Area under the ROC curve for binary protein-interaction classification.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Baselines | MULAN-augmented models are compared with their underlying ESM2 representations.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Leakage controls | Original downstream dataset splits are retained; this alone does not certify protein-pair identity independence or absence from backbone pretraining.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Uncertainty | Downstream results average five initialization seeds; corresponding standard deviations are provided in the supplement.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Entity type | Paper-specific computational evaluation protocol.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Organisms | Human.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Assays | HumanPPI protein-interaction annotations.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Allowed inputs | Protein pairs represented using sequence and structure-informed embeddings.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages |
| Adaptation | Supervised downstream prediction with validation-based tuning/checkpoint selection.SourcesMULAN: multimodal protein language model for sequence and structure encoding · Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; 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 |
|---|---|---|
| MULAN: multimodal protein language model for sequence and structure encoding | journal full text in PMC | Read source DOI: 10.1093/bioadv/vbaf117 |
Historical gaps recorded on 2026-09-17
The catalogue now holds 1 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.
- Complete raw table acquired. HumanPPI AUC column must be extracted separately from localization, thermostability,fluorescence andGO. Different model sizes/representation inputs remain distinct. Structured extraction pending.
Search and extraction details
source found structured extraction pending
Searches
- MULAN: multimodal protein language model for sequence and structure encoding primary paper benchmark results
Evidence locations
- Table2 human-PPI column and supervised task protocols
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.
17 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 | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
Diagram steps
| MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Diagram title Computational evaluation flow Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Datasets HumanPPI downstream task; exact dataset release remains unextracted. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Splits Downstream hyperparameters and checkpoint selection use validation metrics. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Adaptation Supervised downstream prediction with validation-based tuning/checkpoint selection. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Metrics Area under the ROC curve for binary protein-interaction classification. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Baselines MULAN-augmented models are compared with their underlying ESM2 representations. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Leakage controls Original downstream dataset splits are retained; this alone does not certify protein-pair identity independence or absence from backbone pretraining. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. Field: Source artifact SHA-256: Hash scope: Hash scope not separately documented; inspect source record |
| Uncertainty Downstream results average five initialization seeds; corresponding standard deviations are provided in the supplement. Individual claims | MULAN: multimodal protein language model for sequence and structure encoding Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; uncertainty/repeat-run/statistical-comparison passages Version: journal full text in PMC | source checked automated source review · 2026-09-16 Audit detailsTask-specific computational methodology and field context checked in the cited primary-source artifact. Source-backed fields, inapplicable evaluator dimensions and unresolved details are distinguished. Numerical results were not reproduced. 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
- MULAN: multimodal protein language model for sequence and structure encoding · Original source · journal full text in PMC
- MULAN: multimodal protein language model for sequence and structure encoding · Original source · journal full text in PMC
Technical metadata and extraction receipts
Stable ID: reported-task-6e54c7452b2b81
- areas
- proteins-complexes
- tasks
- human protein-protein interaction prediction
- entity level
- task
- version
- Not reported
- task
- human protein-protein interaction prediction
- scope note
- Paper-specific evaluation task; protocol completeness requires further extraction.
- benchmark research
- review date: 2026-09-17; status: source_found_structured_extraction_pending; primary sources: evidence-expansion-p2-mulan-2025-771a9a26ebda; inspected locators: Table2 human-PPI column and supervised task protocols; searched queries: MULAN: multimodal protein language model for sequence and structure encoding primary paper benchmark results; gaps: Complete raw table acquired. HumanPPI AUC column must be extracted separately from localization, thermostability,fluorescence andGO. Different model sizes/representation inputs remain distinct. Structured extraction pending.; 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
- 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: mulan-2025; source locator: Methods §2.5; Results: HumanPPI; cached text lines 31–33, 43; task metric definitions and corresponding results table; 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: HumanPPI
- benchmark: MULAN-ESM2 S: human protein-protein interaction prediction