rewirebio.iobenchmarks
Model

MSA Transformer

Protein language model that takes a multiple sequence alignment as input, interleaving row and column attention, trained by masked language modelling across many protein families; ProteinGym scores it as a zero-shot alignment-based substitution baseline.

4 evaluations · 38 results1 configuration with further results1 open disagreement between studies involves this model

Overview

An explanatory profile has not yet been reviewed for this record.

Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.

Model Benchmark Takeaway

On Livesey and Marsh 2025 VEP rankings, DMS benchmark, all 36 DMS sets (Pairwise win rate, a proxy protocol), MSA_Transformer scored 75.89%, reported by Livesey and Marsh 2025, Supplementary Material 2 (Tables S1-S8) (position 17 of 97 on this protocol for "Use calibrated computational evidence for missense variants (PP3/BP4)"). It also has ranked values in 1 other use case: Predict missense variant effects in a protein of interest.

Recorded parameters: Not recorded. Licence: Not recorded.

Model Capabilities & Hardware Requirements

Deployment facts recorded for MSA Transformer. Fields the record does not carry are marked not recorded.

ParametersNot recorded
Minimum VRAMNot recorded
Inference LatencyNot recorded
Context WindowNot recorded
AccessNot recorded
LicenseNot recorded

Evaluations and results

4 evaluations · 38 results. Different protocols are not a single leaderboard.

Sorted by Spearman (higher is better). The best value in each column is highlighted. Decimals are rounded for display; each value links to the printed value and its source.

Tested configurationProtocol and datasetSpearmanSpearman (ProteinGym zero-shot DMS substitutions, Activity assays: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Binding assays: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Expression assays: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, High MSA depth proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Human proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Low MSA depth proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, mean over selection types of the per-type mean over proteins of assay-level Spearman (all 217 assays))Spearman (ProteinGym zero-shot DMS substitutions, Medium MSA depth proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Organismal fitness assays: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Other eukaryote proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Prokaryote proteins: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Stability assays: mean over proteins of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, variants with 1 substitution: mean over protein and selection-type groups of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, variants with 2 substitutions: mean over protein and selection-type groups of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, variants with 3 substitutions: mean over protein and selection-type groups of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, variants with 4 substitutions: mean over protein and selection-type groups of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, variants with 5 or more substitutions: mean over protein and selection-type groups of assay-level Spearman)Spearman (ProteinGym zero-shot DMS substitutions, Virus proteins: mean over proteins of assay-level Spearman)
MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]EvaluationProteinGym AMFR official zero-shot Spearman snapshot, ProteinGym AMFR 4G3O official snapshot cohort0.24 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
MSA Transformer (single) [ProteinGym 144fe22b snapshot]EvaluationProteinGym AMFR official zero-shot Spearman snapshot, ProteinGym AMFR 4G3O official snapshot cohort0.22Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]EvaluationProteinGym v1.3 zero-shot DMS substitutions, ProteinGym DMS substitution assays at commit 144fe22 (217 assays)Not reported0.473 (best in column)0.329 (best in column)0.446 (best in column)0.479 (best in column)0.439 (best in column)0.375 (best in column)0.432 (best in column)0.456 (best in column)0.419 (best in column)0.516 (best in column)0.446 (best in column)0.492 (best in column)0.427 (best in column)0.221 (best in column)0.365 (best in column)0.369 (best in column)0.397 (best in column)0.421 (best in column)
MSA Transformer (single) [ProteinGym 144fe22b snapshot]EvaluationProteinGym v1.3 zero-shot DMS substitutions, ProteinGym DMS substitution assays at commit 144fe22 (217 assays)Not reported0.4550.3120.4290.4650.430.3640.4170.4410.4120.5010.4260.4750.4120.2080.3630.3610.390.406
All 38 result rows with coverage, uncertainty and sources
Filter evaluations

Applied filters: All linked evaluations

Exact evaluated configurations and original reported results
Tested configurationProtocol and datasetFindingEvidence and details
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym AMFR official zero-shot Spearman snapshot
Dataset subset: ProteinGym AMFR 4G3O official snapshot cohort
0.24 Spearman
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble) AMFR official snapshot

Not reported

Aggregation: Not reported

ProteinGym AMFR assay-level Spearman comparison · DMS_substitutions_Spearman_DMS_level.csv line 15, DMS ID=AMFR_HUMAN_Tsuboyama_2023_4G3O, column=MSA Transformer (ensemble)
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym AMFR official zero-shot Spearman snapshot
Dataset subset: ProteinGym AMFR 4G3O official snapshot cohort
0.22 Spearman
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single) AMFR official snapshot

Not reported

Aggregation: Not reported

ProteinGym AMFR assay-level Spearman comparison · DMS_substitutions_Spearman_DMS_level.csv line 15, DMS ID=AMFR_HUMAN_Tsuboyama_2023_4G3O, column=MSA Transformer (single)
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.432 Spearman (ProteinGym zero-shot DMS substitutions, mean over selection types of the per-type mean over proteins of assay-level Spearman (all 217 assays))
unitless · higher

Uncertainty: SE 0.01. From compute_bootstrap_standard_error_functional_categories in performance_DMS_benchmarks.py at 144fe22 (default 10,000 resamples, lines 95-111 and 310-315); not an absolute standard error of the model's own average.

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Average_Spearman'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.427 Spearman (ProteinGym zero-shot DMS substitutions, variants with 1 substitution: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Depth_1'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.221 Spearman (ProteinGym zero-shot DMS substitutions, variants with 2 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Depth_2'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.365 Spearman (ProteinGym zero-shot DMS substitutions, variants with 3 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Depth_3'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.369 Spearman (ProteinGym zero-shot DMS substitutions, variants with 4 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Depth_4'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.397 Spearman (ProteinGym zero-shot DMS substitutions, variants with 5 or more substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Depth_5+'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.473 Spearman (ProteinGym zero-shot DMS substitutions, Activity assays: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Function_Activity'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.329 Spearman (ProteinGym zero-shot DMS substitutions, Binding assays: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Function_Binding'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.446 Spearman (ProteinGym zero-shot DMS substitutions, Expression assays: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Function_Expression'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.419 Spearman (ProteinGym zero-shot DMS substitutions, Organismal fitness assays: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Function_OrganismalFitness'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.492 Spearman (ProteinGym zero-shot DMS substitutions, Stability assays: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Function_Stability'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.479 Spearman (ProteinGym zero-shot DMS substitutions, High MSA depth proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'High_MSA_depth'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.375 Spearman (ProteinGym zero-shot DMS substitutions, Low MSA depth proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Low_MSA_depth'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.456 Spearman (ProteinGym zero-shot DMS substitutions, Medium MSA depth proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Medium_MSA_depth'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.439 Spearman (ProteinGym zero-shot DMS substitutions, Human proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Taxa_Human'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.516 Spearman (ProteinGym zero-shot DMS substitutions, Other eukaryote proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Taxa_Other_Eukaryote'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.446 Spearman (ProteinGym zero-shot DMS substitutions, Prokaryote proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Taxa_Prokaryote'
Configuration: MSA Transformer (ensemble) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.421 Spearman (ProteinGym zero-shot DMS substitutions, Virus proteins: mean over proteins of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (ensemble): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 37, row 'MSA Transformer (ensemble)', column 'Taxa_Virus'
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.417 Spearman (ProteinGym zero-shot DMS substitutions, mean over selection types of the per-type mean over proteins of assay-level Spearman (all 217 assays))
unitless · higher

Uncertainty: SE 0.01. From compute_bootstrap_standard_error_functional_categories in performance_DMS_benchmarks.py at 144fe22 (default 10,000 resamples, lines 95-111 and 310-315); not an absolute standard error of the model's own average.

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 45, row 'MSA Transformer (single)', column 'Average_Spearman'
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.412 Spearman (ProteinGym zero-shot DMS substitutions, variants with 1 substitution: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 45, row 'MSA Transformer (single)', column 'Depth_1'
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.208 Spearman (ProteinGym zero-shot DMS substitutions, variants with 2 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 45, row 'MSA Transformer (single)', column 'Depth_2'
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.363 Spearman (ProteinGym zero-shot DMS substitutions, variants with 3 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 45, row 'MSA Transformer (single)', column 'Depth_3'
Configuration: MSA Transformer (single) [ProteinGym 144fe22b snapshot]Protocol: ProteinGym v1.3 zero-shot DMS substitutions
Dataset subset: ProteinGym DMS substitution assays at commit 144fe22 (217 assays)
0.361 Spearman (ProteinGym zero-shot DMS substitutions, variants with 4 substitutions: mean over protein and selection-type groups of assay-level Spearman)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Source does not say who ran the evaluation · Source checked
Methods, coverage and source

MSA Transformer (single): ProteinGym zero-shot DMS substitutions summary (144fe22)

ProteinGym zero-shot DMS substitutions, Spearman summary

Aggregation: Not reported

ProteinGym zero-shot DMS substitutions Spearman summary at commit 144fe22 · benchmarks/DMS_zero_shot/substitutions/Spearman/Summary_performance_DMS_substitutions_Spearman.csv line 45, row 'MSA Transformer (single)', column 'Depth_4'

Source checking is not independent reproduction. Release 2026-11-21-adc15377fab2.

Further results on linked records

One linked record has results that are not included above, because no reviewed claim yet ties the link to this model. Each record keeps its own settings and evaluations, and its scores are on its own page.

Use this model

How it works, versions and access

Versions and evaluated configurations

Evidence

Source checking verifies the cited claim or transcription. It does not establish independent reproduction.

Evidence table

Inspect claims, sources and review details

Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.

One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.

3 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-adc15377fab2
Property and statementOriginal source and locationReview and provenance
Description
Protein language model that takes a multiple sequence alignment as input, interleaving row and column attention, trained by masked language modelling across many protein families; ProteinGym scores it as a zero-shot alignment-based substitution baseline.
Individual claims
Abstract: MSA Transformer

Original source ↗

uc20260930-source-proteingym-amfr-spearman: Benchmark CSV header (model scored in the zero-shot DMS substitution benchmark); run-doc-proteingym-readme-md-144fe22b: README 'Leaderboard' baseline table (model name, input modalities, reference); qa-descriptions-20261010-source-msa-transformer: abstract

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: Europe PMC core record PPR/PPR282150 (title and abstract), bioRxiv 2021
Retrieved: 2026-10-10T06:44:38Z

source checked

["automated-source-review","ai-assisted-source-review"] · 2026-10-10

Audit details

Description written from the cited source text; provenance text was generic and is kept here as previous_value. Independent review 2026-10-10: every stated fact is supported by the cited text.

Field: description

Claim: metadata-correction-9abc27593c299dc510f5

Source artifact SHA-256: d1bcaa01d5167374fbb01716d792bd8207c6e974bdd53409486e2bd2ab07a2d0

Hash scope: SHA-256 of one retrieval of a Europe PMC REST search response. The response is not byte-stable: on 2026-10-10 a second retrieval of 10 of these 31 records differed only in JSON key order and parsed to identical content. The title and abstractText fields are the evidence.

Inspected artifact

Description
Protein language model that takes a multiple sequence alignment as input, interleaving row and column attention, trained by masked language modelling across many protein families; ProteinGym scores it as a zero-shot alignment-based substitution baseline.
Individual claims
OATML-Markslab/ProteinGym / README.md

Original source ↗

uc20260930-source-proteingym-amfr-spearman: Benchmark CSV header (model scored in the zero-shot DMS substitution benchmark); run-doc-proteingym-readme-md-144fe22b: README 'Leaderboard' baseline table (model name, input modalities, reference); qa-descriptions-20261010-source-msa-transformer: abstract

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: Git commit 144fe22b07dfaeec2b366f2346203a9838a55b4c
Retrieved: 2026-09-17T09:32:03.517591+00:00

source checked

["automated-source-review","ai-assisted-source-review"] · 2026-10-10

Audit details

Description written from the cited source text; provenance text was generic and is kept here as previous_value. Independent review 2026-10-10: every stated fact is supported by the cited text.

Field: description

Claim: metadata-correction-9abc27593c299dc510f5

Source artifact SHA-256: 321487a8de52c6cfa647a658f61150dd72e0acb0125470524fb94d1f8b23321a

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Description
Protein language model that takes a multiple sequence alignment as input, interleaving row and column attention, trained by masked language modelling across many protein families; ProteinGym scores it as a zero-shot alignment-based substitution baseline.
Individual claims
ProteinGym AMFR assay-level Spearman comparison

Original source ↗

uc20260930-source-proteingym-amfr-spearman: Benchmark CSV header (model scored in the zero-shot DMS substitution benchmark); run-doc-proteingym-readme-md-144fe22b: README 'Leaderboard' baseline table (model name, input modalities, reference); qa-descriptions-20261010-source-msa-transformer: abstract

Shared locator for this statement’s cited sources; not a separate locator for each citation.

Version: 144fe22b07dfaeec2b366f2346203a9838a55b4c
Retrieved: 2026-09-30

source checked

["automated-source-review","ai-assisted-source-review"] · 2026-10-10

Audit details

Description written from the cited source text; provenance text was generic and is kept here as previous_value. Independent review 2026-10-10: every stated fact is supported by the cited text.

Field: description

Claim: metadata-correction-9abc27593c299dc510f5

Source artifact SHA-256: f432423b87f79ac9778dfac86e3d95be041d246bc618dc0a406b35b0b7466437

Hash scope: Hash scope not separately documented; inspect source record

Inspected artifact

Sources and history

Release 2026-11-21-adc15377fab2 · Record review: source checked

3 source records and release historyDownload this release (gzip)
Technical metadata and record identifiers

Technical metadata for this record is tracked under stable ID uc20260930-model-msa-transformer.

Source locator: Pinned ProteinGym Spearman CSV header: MSA Transformer (single)

Related records

How to cite

MSA Transformer. Rewire biological model benchmark database, release 2026-11-21-adc15377fab2. https://benchmarks.rewirebio.io/database/model/uc20260930-model-msa-transformer/

Release 2026-11-21-adc15377fab2. Rankings and values change between releases, so keep the release id in the citation. Cite the original publication as well for any value you reuse: each result links its source.

The compilation and annotations made by Rewire in this release are offered under CC BY 4.0; values and text taken from cited sources stay under the terms of those sources and are not covered by this licence. Data licence

Suggest a correction