Pipeline

CLAPE-SMB with ESM-2

CLAPE-SMB identifies small-molecule-binding residues from protein sequences using frozen ESM-2 features and contrastive learning.

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Conclusion (paragraph 1); Results/Comparison of CLAPE-SMB with DeepProSite and heuristic analyses in three protein case studies (paragraph 3)

1 evaluation · 1 metric row

How it worksEvaluated procedure (conceptual)

Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings.

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5)

At a glance

Model type

Protein sequence transformer; this record is the paper-specific evaluated configuration.

Sources

facebookresearch/esm README.md · README.md model description

Biological inputs

Protein sequences using 20 standard amino-acid tokens and X for unknown residues

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Methods/Sequence embedding (paragraph 2); Conclusion (paragraph 1)

Outputs

Per-residue small-molecule-binding predictions

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Results/CLAPE-SMB can rationally distinguish binding and non-binding sites based on residue features (paragraph 1); Results/Necessity of merging multiple binding sites of similar proteins (paragraph 2)

limited source coverage · Automated source review, 2026-09-16. All specifications and missing details

Evaluations and results

Release 2026-09-17-d277315f7d76 · 1 evaluation · 1 metric row. Different protocols are not a single leaderboard.

Evaluation setupMetricEvidence origin

Results grouped by the exact reported evaluation
Metric and finding	Coverage and uncertainty	Evidence
CLAPE-SMB with ESM-2: protein-small molecule binding-site prediction Pipeline: CLAPE-SMB with ESM-2Task: protein-small molecule binding-site predictionDataset: SJC Contrastive CLAPE-SMB binding-site predictor with ESM-2 feature extractor Author-reported evaluation · Evaluation metadata: needs review
0.917 AUROC Unit: fraction · Direction: unknown	Uncertainty: not reported in legacy extract Scored: Not reported · Eligible: Not reported	source checkedProtein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Table 5, ESM-2 / SJC row, AUROC column Source checking is not independent reproduction.

How it works

How the evaluated method works

The frozen esm2_t33_650M_UR50D encoder emits 1,280-dimensional residue embeddings, which feed the supervised binding-site predictor and contrastive-learning objective.

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5)

Underlying method and version boundaries

ESM-2 is a transformer protein language-model family. The official repository exposes residue embeddings, sequence-level pooling and models at several sizes; the study configuration determines which of these is evaluated.

Sources

facebookresearch/esm README.md · README.md; introduction, model description, pretrained-model and usage sections at pinned revision

What was evaluated

The linked evaluation record identifies CLAPE-SMB with ESM-2: protein-small molecule binding-site prediction. Its dataset, split, adaptation and evidence origin remain attached to the reported results.

Sources

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · The named evaluation’s methods and comparison table; exact preserved evaluation IDs: evaluation-lit-b4-011

Strengths and limitations

Strengths and considerations

Predicts binding sites without requiring an experimentally determined protein structure.
Sources
Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Results/Comparison of CLAPE-SMB with DeepProSite and heuristic analyses in three protein case studies (paragraph 3); Introduction (paragraph 2)

Limitations and conditions

The ESM-2 representation and the supervised binding-site head are jointly part of the evaluated pipeline; reported scores are not raw ESM-2 capabilities.
Sources
Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Results/Computational efficiency of CLAPE-SMB (paragraph 1); Results/Necessity of merging multiple binding sites of similar proteins (paragraph 2)

Profile review details

Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction.

Stable record: reported-model-57dbab30462150

Specifications

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.

Inputs, outputs and configuration
Property	Description and evidence
Model type	Protein sequence transformer; this record is the paper-specific evaluated configuration. Sources facebookresearch/esm README.md · README.md model description
Architecture / procedure	The frozen esm2_t33_650M_UR50D encoder emits 1,280-dimensional residue embeddings, which feed the supervised binding-site predictor and contrastive-learning objective. Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5)
Biological inputs	Protein sequences using 20 standard amino-acid tokens and X for unknown residues Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Methods/Sequence embedding (paragraph 2); Conclusion (paragraph 1)
Outputs	Per-residue small-molecule-binding predictions Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Results/CLAPE-SMB can rationally distinguish binding and non-binding sites based on residue features (paragraph 1); Results/Necessity of merging multiple binding sites of similar proteins (paragraph 2)
Parameters	650-million-parameter, 33-layer ESM-2 feature extractor; total pipeline size is not reported here. Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Table Tab5 (paragraph 1); Methods/Sequence embedding (paragraph 2)
Known versions / configuration	CLAPE-SMB with ESM-2 is the comparison-table label; that label does not specify an immutable weight revision. · Not reported in inspected sources Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label.
Training data / fitting	SJC, UniProtSMB and intrinsically disordered protein datasets described in the paper; encoder layers are not fine-tuned. Sources Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Results/UniProtSMB dataset preparation (paragraph 3); Results/UniProtSMB dataset preparation (paragraph 1)
Context limits	A maximum input/context length for this exact evaluated configuration is not established by the inspected sources. · Not reported in inspected sources Sources (2) Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning; facebookresearch/esm README.md · Methods/Sequence embedding; Methods/Backbone model; Methods/Loss function; Methods/Evaluation metrics; Results/The model architecture of CLAPE-SMB; Results/Influence of model architecture on prediction accuracy; inspected for explicit maximum input length (dataset lengths and family-wide limits are not substituted); README.md at pinned repository revision
Access	Official upstream implementation and usage documentation: https://github.com/facebookresearch/esm/blob/2b369911bb5b4b0dda914521b9475cad1656b2ac/README.md. This pinned documentation revision is not automatically the evaluated weight revision. Sources facebookresearch/esm README.md · README.md; installation, model download and usage instructions
Code licence	MIT (upstream repository code at the cited revision; this does not establish every dependency or historical checkpoint licence). Sources facebookresearch/esm LICENSE · LICENSE; complete licence text
Weights licence	The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. · Not reported in inspected sources Sources facebookresearch/esm README.md · README.md; checkpoint/access documentation and licence scope

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.

21 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-09-17-d277315f7d76
Property and statement	Original source and location	Review and provenance
Diagram caption Conceptual input–method–output guide. Check the procedure text and linked evaluation for fitted components, additional inputs and exact settings. Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.diagram.caption` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram steps ["Protein sequences using 20 standard amino-acid tokens and X for unknown residues","CLAPE-SMB with ESM-2","Per-residue small-molecule-binding predictions"] Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.diagram.steps` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Diagram title Evaluated procedure (conceptual) Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.diagram.title` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Model type Protein sequence transformer; this record is the paper-specific evaluated configuration. Individual claims	facebookresearch/esm README.md Original source ↗ README.md model description Version: 2b369911bb5b4b0dda914521b9475cad1656b2ac Retrieved: 2026-09-16T20:00:00.816433+00:00	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.0.value` Source artifact SHA-256: `8b273c21a322fc9473d1b68d0dd40c8166ab2f89e4a190aa26ca87251b97cba9` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Architecture / procedure The frozen esm2_t33_650M_UR50D encoder emits 1,280-dimensional residue embeddings, which feed the supervised binding-site predictor and contrastive-learning objective. Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.1.value` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Weights licence The inspected model-access documentation does not explicitly identify terms for this exact evaluated checkpoint or fitted head; repository code terms are shown separately. Individual claims	facebookresearch/esm README.md Original source ↗ README.md; checkpoint/access documentation and licence scope Version: 2b369911bb5b4b0dda914521b9475cad1656b2ac Retrieved: 2026-09-16T20:00:00.816433+00:00	unreported automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.10.value` Source artifact SHA-256: `8b273c21a322fc9473d1b68d0dd40c8166ab2f89e4a190aa26ca87251b97cba9` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Biological inputs Protein sequences using 20 standard amino-acid tokens and X for unknown residues Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Methods/Sequence embedding (paragraph 2); Conclusion (paragraph 1) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.2.value` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Outputs Per-residue small-molecule-binding predictions Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Results/CLAPE-SMB can rationally distinguish binding and non-binding sites based on residue features (paragraph 1); Results/Necessity of merging multiple binding sites of similar proteins (paragraph 2) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.3.value` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Parameters 650-million-parameter, 33-layer ESM-2 feature extractor; total pipeline size is not reported here. Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Table Tab5 (paragraph 1); Methods/Sequence embedding (paragraph 2) Version: version of record Retrieved: 2026-09-16T10:41:06Z	source checked automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.4.value` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact
Known versions / configuration CLAPE-SMB with ESM-2 is the comparison-table label; that label does not specify an immutable weight revision. Individual claims	Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning Original source ↗ Model identification in the comparison table and corresponding Methods; immutable checkpoint revision is not supplied by the table label. Version: version of record Retrieved: 2026-09-16T10:41:06Z	unreported automated source review · 2026-09-16 Audit details Primary full text and the available official implementation/model documentation were inspected. Explanatory claims are source-backed; unresolved exact-configuration metadata is labelled explicitly. This is automated review, not a human review or independent benchmark reproduction. Field: `attributes.profile.facts.5.value` Source artifact SHA-256: `215919244c3dd2dfb0b55fce91c211430fd8d4aee4bb28bd03eab9f4feb73e62` Hash scope: Hash scope not separately documented; inspect source record Inspected artifact

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

Sources and history

Release 2026-09-17-d277315f7d76 · Record review: needs review

3 source records and release history

Protein-small molecule binding site prediction based on a pre-trained protein language model with contrastive learning · Original source · version of record
facebookresearch/esm README.md · Original source · 2b369911bb5b4b0dda914521b9475cad1656b2ac
facebookresearch/esm LICENSE · Original source · 2b369911bb5b4b0dda914521b9475cad1656b2ac

Download this release

Technical metadata and extraction receipts

Stable ID: reported-model-57dbab30462150

areas: proteins-complexes
entity level: method
version: Not reported
reported name: CLAPE-SMB with ESM-2
historical missing metadata: version: not_reported_in_legacy_extract; checkpoint revision: not_reported_in_legacy_extract; training data: not_reported_in_legacy_extract; licence: 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: model
entity classification: review date: 2026-09-17; rationale: This record identifies a composed analysis workflow with separately identifiable upstream models, representations or tools and a downstream prediction/scoring procedure. Results belong to that complete composition rather than to an upstream model alone.; source ids: clape-smb-2024; evidence-reported-base-esm-readme-md; source locator: Methods/Sequence embedding (paragraph 2); Methods/Loss function (paragraph 5) | README.md model description | Conclusion (paragraph 1); Results/Comparison of CLAPE-SMB with DeepProSite and heuristic analyses in three protein case studies (paragraph 3); ambiguities: This is the paper-specific pipeline identity; unspecified component checkpoints or implementation versions are not inferred from its name.

Related records

model: CLAPE-SMB with ESM-2: protein-small molecule binding-site prediction

Suggest a correction