rewirebio.iobenchmarks
Configuration

LigandMPNN (HyInv report)

LigandMPNN decoding autoregressively at temperature 0.1 on backbone coordinates only.

40 evaluations · 40 results

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 Capabilities & Hardware Requirements

Deployment facts recorded for LigandMPNN (HyInv report). 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

Evaluated Checkpoint: Released weights at the published setting, as run in the HyInv technical report; no checkpoint revision printed

Evaluations and results

40 evaluations · 40 results. Different protocols are not a single leaderboard.

Sorted by lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1) (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 datasetlDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)Sctm (Mean TM-score of the predicted structure to the native backbone, averaged over the designs of a target and then over targets (macro-average); computed on the CA trace at valid sites, normalised by the native length and clipped at one; ESMFold2 as the structure predictor, which is also the reinforcement-learning reward of HyInv)Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)Sequence recovery (Fraction of valid sites at which the design carries the native residue, averaged over the designs of a target and then over targets; descriptive in this report, with weight zero in HyInv's reward and no role in checkpoint selection)Runtime (Median wall-clock time of one design request on one NVIDIA H20, 128 candidates per request with a maximum candidate batch of 32 unless the request is the 10,000-design library; the timed region excludes model loading, CUDA graph capture and kernel compilation; the backbone length is named per result; the request is 128 designs for a 402-residue backbone)Sctm (Mean TM-score of the predicted structure to the native backbone, averaged over the designs of a target and then over targets; Boltz2 as the structure predictor, with the alignment built from the designed sequence by MMseqs2 against UniRef50; RecentPDB only)Sctm (Mean TM-score of the predicted structure to the native backbone, averaged over the designs of a target and then over targets; Protenix2 as the structure predictor, with the alignment built from the designed sequence by MMseqs2 against UniRef50; RecentPDB only)Sctm difference (HyInv default minus the baseline, paired by chain on RecentPDB with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; Boltz2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)Sctm difference (HyInv default minus the baseline, paired by chain on RecentPDB with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; Protenix2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)Sequence diversity (Mean pairwise normalised Hamming distance between the K designs of a chain over valid sites, then a macro-average over targets; counted at valid sites for every method, because baseline sequences carry X at invalid sites and counting those as identical would lower their diversity)
LigandMPNN (HyInv report)EvaluationCA-lDDT of designed sequences under ESMFold2 (HyInv report), TS500 (500 chains)0.896 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationCA-lDDT of designed sequences under ESMFold2 (HyInv report), TS50 (50 chains)0.891Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationCA-lDDT of designed sequences under ESMFold2 (HyInv report), CATH4.2 test (1,120 chains)0.862Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationCA-lDDT of designed sequences under ESMFold2 (HyInv report), CATH4.3 test (1,882 chains)0.859Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationCA-lDDT of designed sequences under ESMFold2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)0.852Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationLatency of a design request on one NVIDIA H20 (HyInv report), Timing backbones for the HyInv inference-time measurementsNot reportedNot reportedNot reportedNot reportedNot reportedNot reported79 sNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationpLDDT of ESMFold2 predictions of designed sequences (HyInv report), CATH4.2 test (1,120 chains)Not reported81.6Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationpLDDT of ESMFold2 predictions of designed sequences (HyInv report), CATH4.3 test (1,882 chains)Not reported81.2Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationpLDDT of ESMFold2 predictions of designed sequences (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reported83.5Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationpLDDT of ESMFold2 predictions of designed sequences (HyInv report), TS50 (50 chains)Not reported84.6Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationpLDDT of ESMFold2 predictions of designed sequences (HyInv report), TS500 (500 chains)Not reported89.8 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSelf-consistency RMSD of designed sequences under ESMFold2 (HyInv report), CATH4.2 test (1,120 chains)Not reportedNot reported4.03Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSelf-consistency RMSD of designed sequences under ESMFold2 (HyInv report), CATH4.3 test (1,882 chains)Not reportedNot reported4.24Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSelf-consistency RMSD of designed sequences under ESMFold2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reported5.21Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSelf-consistency RMSD of designed sequences under ESMFold2 (HyInv report), TS50 (50 chains)Not reportedNot reported2.48 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSelf-consistency RMSD of designed sequences under ESMFold2 (HyInv report), TS500 (500 chains)Not reportedNot reported4.04Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under Boltz2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.854Not reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under Boltz2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported+20.1 [8.4, 33.0]Not reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under ESMFold2 (HyInv report), CATH4.2 test (1,120 chains)Not reportedNot reportedNot reportedNot reported+14.7 [6.6, 26.1]Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under ESMFold2 (HyInv report), CATH4.3 test (1,882 chains)Not reportedNot reportedNot reportedNot reported+16.6 [8.1, 28.1]Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under ESMFold2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reported+31.6 [18.2, 46.2] (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under ESMFold2 (HyInv report), TS50 (50 chains)Not reportedNot reportedNot reportedNot reported+8.7 [−5.2, 23.5]Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under ESMFold2 (HyInv report), TS500 (500 chains)Not reportedNot reportedNot reportedNot reported+9.4 [4.4, 14.9]Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationPaired scTM difference from HyInv under Protenix2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported+22.6 [8.7, 37.1]Not reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under ESMFold2 (HyInv report), CATH4.2 test (1,120 chains)Not reportedNot reportedNot reported0.841Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under ESMFold2 (HyInv report), CATH4.3 test (1,882 chains)Not reportedNot reportedNot reported0.823Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under ESMFold2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reported0.829Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under ESMFold2 (HyInv report), TS50 (50 chains)Not reportedNot reportedNot reported0.912 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under ESMFold2 (HyInv report), TS500 (500 chains)Not reportedNot reportedNot reported0.889Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationscTM of designed sequences under Protenix2 (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.856Not reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationSequence diversity of a design set (HyInv report), CATH4.2 test (1,120 chains)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.272
LigandMPNN (HyInv report)EvaluationSequence diversity of a design set (HyInv report), CATH4.3 test (1,882 chains)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.277
LigandMPNN (HyInv report)EvaluationSequence diversity of a design set (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.266
LigandMPNN (HyInv report)EvaluationSequence diversity of a design set (HyInv report), TS50 (50 chains)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.224
LigandMPNN (HyInv report)EvaluationSequence diversity of a design set (HyInv report), TS500 (500 chains)Not reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reportedNot reported0.22
LigandMPNN (HyInv report)EvaluationNative sequence recovery of designed sequences (HyInv report), CATH4.2 test (1,120 chains)Not reportedNot reportedNot reportedNot reportedNot reported0.427Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationNative sequence recovery of designed sequences (HyInv report), CATH4.3 test (1,882 chains)Not reportedNot reportedNot reportedNot reportedNot reported0.411Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationNative sequence recovery of designed sequences (HyInv report), RecentPDB (121 chains from 111 recent PDB entries)Not reportedNot reportedNot reportedNot reportedNot reported0.453Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationNative sequence recovery of designed sequences (HyInv report), TS50 (50 chains)Not reportedNot reportedNot reportedNot reportedNot reported0.475Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
LigandMPNN (HyInv report)EvaluationNative sequence recovery of designed sequences (HyInv report), TS500 (500 chains)Not reportedNot reportedNot reportedNot reportedNot reported0.511 (best in column)Not reportedNot reportedNot reportedNot reportedNot reportedNot reported
All 40 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: LigandMPNN (HyInv report)Protocol: Latency of a design request on one NVIDIA H20 (HyInv report)
Dataset: Timing backbones for the HyInv inference-time measurements
79 s Runtime (Median wall-clock time of one design request on one NVIDIA H20, 128 candidates per request with a maximum candidate batch of 32 unless the request is the 10,000-design library; the timed region excludes model loading, CUDA graph capture and kernel compilation; the backbone length is named per result; the request is 128 designs for a 402-residue backbone)
second · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

Latency of a design request for ligandmpnn on one NVIDIA H20 (HyInv report)

hyinv-20261106-protocol-design-latency-h20

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 4a, bar 'LigandMPNN', printed label (page 5 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: CA-lDDT of designed sequences under ESMFold2 (HyInv report)
Dataset: CATH4.2 test (1,120 chains)
0.862 lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.2 test (CA-lDDT of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-lddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'CA-lDDT' block, row 'LigandMPNN', column 'CATH4.2 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: CA-lDDT of designed sequences under ESMFold2 (HyInv report)
Dataset: CATH4.3 test (1,882 chains)
0.859 lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.3 test (CA-lDDT of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-lddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'CA-lDDT' block, row 'LigandMPNN', column 'CATH4.3 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: CA-lDDT of designed sequences under ESMFold2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
0.852 lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on RecentPDB (CA-lDDT of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-lddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'CA-lDDT' block, row 'LigandMPNN', column 'RecentPDB' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: CA-lDDT of designed sequences under ESMFold2 (HyInv report)
Dataset: TS50 (50 chains)
0.891 lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS50 (CA-lDDT of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-lddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'CA-lDDT' block, row 'LigandMPNN', column 'TS50' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: CA-lDDT of designed sequences under ESMFold2 (HyInv report)
Dataset: TS500 (500 chains)
0.896 lDDT (CA-only lDDT-style score without superposition, over all CA pairs below 15 Angstrom in the native structure, as the fraction whose distance error is below 0.5, 1, 2 and 4 Angstrom averaged over the four thresholds with the pairs pooled over the whole design; between 0 and 1)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS500 (CA-lDDT of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-lddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'CA-lDDT' block, row 'LigandMPNN', column 'TS500' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: pLDDT of ESMFold2 predictions of designed sequences (HyInv report)
Dataset: CATH4.2 test (1,120 chains)
81.6 Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.2 test (pLDDT of ESMFold2 predictions of designed sequences, HyInv report)

hyinv-20261106-protocol-plddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'pLDDT' block, row 'LigandMPNN', column 'CATH4.2 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: pLDDT of ESMFold2 predictions of designed sequences (HyInv report)
Dataset: CATH4.3 test (1,882 chains)
81.2 Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.3 test (pLDDT of ESMFold2 predictions of designed sequences, HyInv report)

hyinv-20261106-protocol-plddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'pLDDT' block, row 'LigandMPNN', column 'CATH4.3 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: pLDDT of ESMFold2 predictions of designed sequences (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
83.5 Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on RecentPDB (pLDDT of ESMFold2 predictions of designed sequences, HyInv report)

hyinv-20261106-protocol-plddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'pLDDT' block, row 'LigandMPNN', column 'RecentPDB' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: pLDDT of ESMFold2 predictions of designed sequences (HyInv report)
Dataset: TS50 (50 chains)
84.6 Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS50 (pLDDT of ESMFold2 predictions of designed sequences, HyInv report)

hyinv-20261106-protocol-plddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'pLDDT' block, row 'LigandMPNN', column 'TS50' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: pLDDT of ESMFold2 predictions of designed sequences (HyInv report)
Dataset: TS500 (500 chains)
89.8 Plddt (ESMFold2 mean per-residue confidence over valid sites on a scale from 0 to 100, averaged over the designs of a target and then over targets; from a second fold of the same sequences with the confidence head on)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS500 (pLDDT of ESMFold2 predictions of designed sequences, HyInv report)

hyinv-20261106-protocol-plddt-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'pLDDT' block, row 'LigandMPNN', column 'TS500' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Self-consistency RMSD of designed sequences under ESMFold2 (HyInv report)
Dataset: CATH4.2 test (1,120 chains)
4.03 Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)
angstrom · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.2 test (Self-consistency RMSD of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-scrmsd-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'scRMSD' block, row 'LigandMPNN', column 'CATH4.2 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Self-consistency RMSD of designed sequences under ESMFold2 (HyInv report)
Dataset: CATH4.3 test (1,882 chains)
4.24 Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)
angstrom · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.3 test (Self-consistency RMSD of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-scrmsd-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'scRMSD' block, row 'LigandMPNN', column 'CATH4.3 test' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Self-consistency RMSD of designed sequences under ESMFold2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
5.21 Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)
angstrom · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on RecentPDB (Self-consistency RMSD of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-scrmsd-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'scRMSD' block, row 'LigandMPNN', column 'RecentPDB' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Self-consistency RMSD of designed sequences under ESMFold2 (HyInv report)
Dataset: TS50 (50 chains)
2.48 Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)
angstrom · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS50 (Self-consistency RMSD of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-scrmsd-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'scRMSD' block, row 'LigandMPNN', column 'TS50' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Self-consistency RMSD of designed sequences under ESMFold2 (HyInv report)
Dataset: TS500 (500 chains)
4.04 Scrmsd (CA RMSD between the ESMFold2 prediction of a design and the native backbone at valid sites after optimal Kabsch superposition with the residue correspondence fixed by position, averaged over the designs of a target and then over targets)
angstrom · lower

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on TS500 (Self-consistency RMSD of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-scrmsd-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Figure 15, 'scRMSD' block, row 'LigandMPNN', column 'TS500' (page 17 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: scTM of designed sequences under Boltz2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
0.854 Sctm (Mean TM-score of the predicted structure to the native backbone, averaged over the designs of a target and then over targets; Boltz2 as the structure predictor, with the alignment built from the designed sequence by MMseqs2 against UniRef50; RecentPDB only)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on RecentPDB Boltz2 (scTM of designed sequences under Boltz2, HyInv report)

hyinv-20261106-protocol-sctm-boltz2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 6, row 'LigandMPNN', column 'RecentPDB, Boltz2' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under Boltz2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
+20.1 [8.4, 33.0] Sctm difference (HyInv default minus the baseline, paired by chain on RecentPDB with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; Boltz2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 8.4 to 33.0. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on RecentPDB Boltz2 (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-boltz2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'RecentPDB, Boltz2' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under ESMFold2 (HyInv report)
Dataset: CATH4.2 test (1,120 chains)
+14.7 [6.6, 26.1] Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 6.6 to 26.1. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on CATH4.2 test (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'CATH4.2 test' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under ESMFold2 (HyInv report)
Dataset: CATH4.3 test (1,882 chains)
+16.6 [8.1, 28.1] Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 8.1 to 28.1. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on CATH4.3 test (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'CATH4.3 test' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under ESMFold2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
+31.6 [18.2, 46.2] Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 18.2 to 46.2. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on RecentPDB (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'RecentPDB, ESMFold2' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under ESMFold2 (HyInv report)
Dataset: TS50 (50 chains)
+8.7 [−5.2, 23.5] Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI -5.2 to 23.5. Printed in units of 10^-3 scTM, as the point estimate is. The interval includes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on TS50 (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'TS50' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under ESMFold2 (HyInv report)
Dataset: TS500 (500 chains)
+9.4 [4.4, 14.9] Sctm difference (HyInv default minus the baseline, paired by chain on the same panel with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; ESMFold2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 4.4 to 14.9. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on TS500 (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'TS500' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: Paired scTM difference from HyInv under Protenix2 (HyInv report)
Dataset: RecentPDB (121 chains from 111 recent PDB entries)
+22.6 [8.7, 37.1] Sctm difference (HyInv default minus the baseline, paired by chain on RecentPDB with shared folding randomness, printed in units of 10^-3 scTM with a 95% percentile interval; Protenix2 as the structure predictor; the reference system is the HyInv default setting; the baseline is LigandMPNN)
unitless · higher

Uncertainty: 95% CI 8.7 to 37.1. Printed in units of 10^-3 scTM, as the point estimate is. The interval excludes zero.

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

HyInv minus LigandMPNN on RecentPDB Protenix2 (paired scTM difference, HyInv report)

hyinv-20261106-protocol-sctm-difference-protenix2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 7, row 'LigandMPNN', column 'RecentPDB, Protenix2' (page 16 of 21)
Configuration: LigandMPNN (HyInv report)Protocol: scTM of designed sequences under ESMFold2 (HyInv report)
Dataset: CATH4.2 test (1,120 chains)
0.841 Sctm (Mean TM-score of the predicted structure to the native backbone, averaged over the designs of a target and then over targets (macro-average); computed on the CA trace at valid sites, normalised by the native length and clipped at one; ESMFold2 as the structure predictor, which is also the reinforcement-learning reward of HyInv)
unitless · higher

Uncertainty: Not reported by the source

Coverage: Not reported scored / Not reported eligible

Author-reported evaluation · Source checked
Methods, coverage and source

LigandMPNN on CATH4.2 test (scTM of designed sequences under ESMFold2, HyInv report)

hyinv-20261106-protocol-sctm-esmfold2

Aggregation: Not reported

HyInv Technical Report (Hyra Team, 2026-10-14) · Table 6, row 'LigandMPNN', column 'CATH4.2 test' (page 16 of 21)

Source checking is not independent reproduction. Release 2026-11-21-c8bea25e5abb.

Use this configuration

How it works, versions and access

Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this configuration. Consult the linked sources and evaluation records below.

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.

0 evidence rows matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-c8bea25e5abb
Property and statementOriginal source and locationReview and provenance

No evidence rows match these filters. Choose another scope or clear the search.

Sources and history

Release 2026-11-21-c8bea25e5abb · Record review: source checked

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

Technical metadata for this record is tracked under stable ID hyinv-20261106-config-ligandmpnn.

Source locator: Table 5 row 'LigandMPNN'; Section 5.1 'Baselines'; Appendix D.1

Related records

How to cite

LigandMPNN (HyInv report). Rewire biological model benchmark database, release 2026-11-21-c8bea25e5abb. https://benchmarks.rewirebio.io/database/configuration/hyinv-20261106-config-ligandmpnn/

Release 2026-11-21-c8bea25e5abb. 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