ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
Zero-shot substitutions, Spearman. Scored with Spearman on ProteinGym substitution DMS assays. Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes.
Overview
Zero-shot substitutions, Spearman. Scored with Spearman on ProteinGym substitution DMS assays. Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Benchmarks
These source-backed links do not make different protocols or scores interchangeable.
Recorded evaluations
Each evaluation records what was tested and under which conditions.
- CARP (640M) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- DeepSequence (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- ESM-1b on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- ESM-1v (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- ESM-2 (15B) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- ESM-IF1 on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- EVE (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- EVmutation on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- GEMME on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- MIF-ST on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- MSA Transformer (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- ProGen2 XL on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
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.
Published comparisons
Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.
ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
spearman (correlation) · Higher values are better for this metric.
Every method ProteinGym reports on Zero-shot substitutions, Spearman, scored with Spearman on ProteinGym substitution DMS assays.
Evaluation protocol · ProteinGym substitution DMS assays (ProteinGym split)
- Site-Independent · Configuration · Author-reported evaluation0.359
- WaveNet · Configuration · Author-reported evaluation0.373
- EVmutation · Configuration · Author-reported evaluation0.395
- DeepSequence (ensemble) · Configuration · Author-reported evaluation0.419
- EVE (ensemble) · Configuration · Author-reported evaluation0.439
- GEMME · Configuration · Author-reported evaluation0.455
- UniRep · Configuration · Author-reported evaluation0.190
- CARP (640M) · Configuration · Author-reported evaluation0.368
- ESM-1b · Configuration · Author-reported evaluation0.394
- ESM-2 (15B) · Configuration · Author-reported evaluation0.401
- RITA XL · Configuration · Author-reported evaluation0.372
- ESM-1v (ensemble) · Configuration · Author-reported evaluation0.407
- ProGen2 XL · Configuration · Author-reported evaluation0.391
- VESPA · Configuration · Author-reported evaluation0.436
- UniRep evotuned · Configuration · Author-reported evaluation0.347
- MSA Transformer (ensemble) · Configuration · Author-reported evaluation0.434
- Tranception L · Configuration · Author-reported evaluation0.434
- TranceptEVE L · Configuration · Author-reported evaluation0.456
- ESM-IF1 · Configuration · Author-reported evaluation0.422
- MIF-ST · Configuration · Author-reported evaluation0.401
- ProteinMPNN · Configuration · Author-reported evaluation0.258
Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.
ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, column(Zero-shot substitutions, Spearman)Values, uncertainty and evidence
| Tested entity | Printed value | Uncertainty | Evidence |
|---|---|---|---|
| Site-Independent · Configuration | 0.359 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Alignment- Site-Independent), column(Zero-shot substitutions, Spearman) |
| WaveNet · Configuration | 0.373 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(based WaveNet), column(Zero-shot substitutions, Spearman) |
| EVmutation · Configuration | 0.395 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(EVmutation), column(Zero-shot substitutions, Spearman) |
| DeepSequence (ensemble) · Configuration | 0.419 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(DeepSequence (ensemble)), column(Zero-shot substitutions, Spearman) |
| EVE (ensemble) · Configuration | 0.439 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(EVE (ensemble)), column(Zero-shot substitutions, Spearman) |
| GEMME · Configuration | 0.455 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(GEMME), column(Zero-shot substitutions, Spearman) |
| UniRep · Configuration | 0.190 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Protein UniRep), column(Zero-shot substitutions, Spearman) |
| CARP (640M) · Configuration | 0.368 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(language CARP (640M)), column(Zero-shot substitutions, Spearman) |
| ESM-1b · Configuration | 0.394 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1b), column(Zero-shot substitutions, Spearman) |
| ESM-2 (15B) · Configuration | 0.401 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-2 (15B)), column(Zero-shot substitutions, Spearman) |
| RITA XL · Configuration | 0.372 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(RITA XL), column(Zero-shot substitutions, Spearman) |
| ESM-1v (ensemble) · Configuration | 0.407 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, Spearman) |
| ProGen2 XL · Configuration | 0.391 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ProGen2 XL), column(Zero-shot substitutions, Spearman) |
| VESPA · Configuration | 0.436 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(VESPA), column(Zero-shot substitutions, Spearman) |
| UniRep evotuned · Configuration | 0.347 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Hybrid UniRep evotuned), column(Zero-shot substitutions, Spearman) |
| MSA Transformer (ensemble) · Configuration | 0.434 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(MSA Transformer (ensemble)), column(Zero-shot substitutions, Spearman) |
| Tranception L · Configuration | 0.434 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Tranception L), column(Zero-shot substitutions, Spearman) |
| TranceptEVE L · Configuration | 0.456 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(TranceptEVE L), column(Zero-shot substitutions, Spearman) |
| ESM-IF1 · Configuration | 0.422 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Inverse ESM-IF1), column(Zero-shot substitutions, Spearman) |
| MIF-ST · Configuration | 0.401 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Folding MIF-ST), column(Zero-shot substitutions, Spearman) |
| ProteinMPNN · Configuration | 0.258 correlation | Not reported | Author-reported evaluation · source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ProteinMPNN), column(Zero-shot substitutions, Spearman) |
Scope and limitations
- Author-reported numbers, source checked but not independently reproduced.
- MSE is better when lower; every other metric here is better when higher.
- Zero-shot, supervised and indel benchmarks use different protocols and assay sets, so their figures are not comparable to each other.
Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.
Tested entities and results
Release 2026-09-17-134cd1815de8 · 21 evaluations · 21 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| CARP (640M) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: CARP (640M)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.368 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(language CARP (640M)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| DeepSequence (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: DeepSequence (ensemble)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.419 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(DeepSequence (ensemble)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ESM-1b on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ESM-1bTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.394 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1b), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ESM-1v (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ESM-1v (ensemble)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.407 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-1v (ensemble)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ESM-2 (15B) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ESM-2 (15B)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.401 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ESM-2 (15B)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ESM-IF1 on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ESM-IF1Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.422 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Inverse ESM-IF1), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| EVE (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: EVE (ensemble)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.439 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(EVE (ensemble)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| EVmutation on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: EVmutationTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.395 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(EVmutation), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| GEMME on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: GEMMETask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.455 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(GEMME), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| MIF-ST on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: MIF-STTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.401 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Folding MIF-ST), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| MSA Transformer (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: MSA Transformer (ensemble)Task: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.434 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(MSA Transformer (ensemble)), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ProGen2 XL on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ProGen2 XLTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.391 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ProGen2 XL), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| ProteinMPNN on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: ProteinMPNNTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.258 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(ProteinMPNN), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| RITA XL on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: RITA XLTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.372 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(RITA XL), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| Site-Independent on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: Site-IndependentTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.359 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Alignment- Site-Independent), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| TranceptEVE L on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: TranceptEVE LTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.456 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(TranceptEVE L), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| Tranception L on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: Tranception LTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.434 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Tranception L), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| UniRep evotuned on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: UniRep evotunedTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.347 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Hybrid UniRep evotuned), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| UniRep on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: UniRepTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.190 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(Protein UniRep), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| VESPA on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: VESPATask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.436 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(VESPA), column(Zero-shot substitutions, Spearman) Source checking is not independent reproduction. |
| WaveNet on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman Configuration: WaveNetTask: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, SpearmanDataset subset: ProteinGym substitution DMS assays (ProteinGym split) Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.373 spearman Unit: correlation · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Table 2, row(based WaveNet), column(Zero-shot substitutions, Spearman) Source checking is not 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.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: part of discovery-benchmark-proteingym Individual claims | ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design Table 2, column(Zero-shot substitutions, Spearman) Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256 | source checked automated source review · 2026-09-18 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: proteingym-association-zs-sub-spearman Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-09-17-134cd1815de8 · Record review: source checked
1 source records and release history
- ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Technical metadata and extraction receipts
Stable ID: proteingym-task-zs-sub-spearman
- areas
- proteins-complexes
- tasks
- Zero-shot substitutions, Spearman
- metric
- Spearman
- metric direction
- higher
- dataset
- ProteinGym substitution DMS assays
- protocol
- Zero-shot scoring of substitution assays, averaged over assays with the correction the paper describes.
- source locator
- Table 2, column(Zero-shot substitutions, Spearman)
- comparison panels
- id: proteingym-panel-zs-sub-spearman; title: ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman; protocol id: proteingym-task-zs-sub-spearman; dataset id: proteingym-dataset-proteingym-substitution-dms-assays; metric: spearman; unit: correlation; direction: higher; result ids: proteingym-result-site-independent-zs-sub-spearman-spearman; proteingym-result-wavenet-zs-sub-spearman-spearman; proteingym-result-evmutation-zs-sub-spearman-spearman; proteingym-result-deepsequence-ensemble-zs-sub-spearman-spearman; proteingym-result-eve-ensemble-zs-sub-spearman-spearman; proteingym-result-gemme-zs-sub-spearman-spearman; proteingym-result-unirep-zs-sub-spearman-spearman; proteingym-result-carp-640m-zs-sub-spearman-spearman; proteingym-result-esm-1b-zs-sub-spearman-spearman; proteingym-result-esm-2-15b-zs-sub-spearman-spearman; proteingym-result-rita-xl-zs-sub-spearman-spearman; proteingym-result-esm-1v-ensemble-zs-sub-spearman-spearman; proteingym-result-progen2-xl-zs-sub-spearman-spearman; proteingym-result-vespa-zs-sub-spearman-spearman; proteingym-result-unirep-evotuned-zs-sub-spearman-spearman; proteingym-result-msa-transformer-ensemble-zs-sub-spearman-spearman; proteingym-result-tranception-l-zs-sub-spearman-spearman; proteingym-result-trancepteve-l-zs-sub-spearman-spearman; proteingym-result-esm-if1-zs-sub-spearman-spearman; proteingym-result-mif-st-zs-sub-spearman-spearman; proteingym-result-proteinmpnn-zs-sub-spearman-spearman; source ids: evidence-expansion-proteingym-a3b08cc4; source locator: Table 2, column(Zero-shot substitutions, Spearman); context: Every method ProteinGym reports on Zero-shot substitutions, Spearman, scored with Spearman on ProteinGym substitution DMS assays.; caveats: Author-reported numbers, source checked but not independently reproduced.; MSE is better when lower; every other metric here is better when higher.; Zero-shot, supervised and indel benchmarks use different protocols and assay sets, so their figures are not comparable to each other.; review: method: automated_source_review; date: 2026-09-18
Related records
- part of: ProteinGym
- subject: ProteinGym ZS-SUB-SPEARMAN: part of discovery-benchmark-proteingym
- benchmark: CARP (640M) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: DeepSequence (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ESM-1b on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ESM-1v (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ESM-2 (15B) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ESM-IF1 on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: EVE (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: EVmutation on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: GEMME on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: MIF-ST on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: MSA Transformer (ensemble) on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ProGen2 XL on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: ProteinMPNN on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: RITA XL on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: Site-Independent on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: TranceptEVE L on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: Tranception L on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: UniRep evotuned on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: UniRep on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: VESPA on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman
- benchmark: WaveNet on ProteinGym ZS-SUB-SPEARMAN: Zero-shot substitutions, Spearman