| Comparison: adaptation HyInv runs its default setting. ProteinMPNN and ProteinZero V2 are interpolated at the HyInv diversity from their temperature grids, without extrapolating by more than 0.05, so the baseline side is interpolated rather than measured at HyInv's diversity Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.adaptation Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: aggregation Mean number of qualifying designs per backbone over the 74 backbones; the resampling unit is the cluster of 30% identity. Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.aggregation Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: budget 4 designs per backbone Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.budget Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: dataset version As assembled for the HyInv technical report (2026-10-14); no chain list or split manifest is published Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.dataset_version Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: inputs Backbone N, CA, C and O coordinates of the panel chain; no sequence, alignment, language model or structure predictor at design time Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.inputs Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: metric implementation 74 de novo backbones, 4 designs each. A design is foldable if its ESMFold2 scTM is at least 0.9 and novel if its identity to the nearest UniRef50 sequence and to the nearest pretraining-library sequence is below 0.3. The composition filter keeps sequences with entropy at least 3.5 bits, hydrophobic fraction (AVILMFWY) between 0.25 and 0.55 and longest homopolymer run at most 6, which excludes low-complexity sequences a structure predictor can fold to a high scTM. Both readings were fixed before designs were generated. HyInv is compared with ProteinMPNN and ProteinZero V2 over their temperature grids, interpolating each curve at the HyInv diversity without extrapolating by more than 0.05. Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.metric_implementation Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: population 74 de novo backbones in 58 clusters of 30% sequence identity, none with a hit that passes the report's homology rule; 28 have partial hits below the coverage limit (Table 10) Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.population Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: protocol id hyinv-20261106-protocol-novel-and-foldable-yield Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.protocol_id Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Comparison: split application panel Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.comparison.split Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|
| Origin author_reported Context-only references | HyInv Technical Report (Hyra Team, 2026-10-14) Original source ↗ Section 6 'De novo backbones', sentence '0.68 unfiltered and 0.38 filtered per backbone, against 0.47 and 0.05 for ProteinMPNN and 0.45 and 0.11 for ProteinZero V2' (page 8 of 21) Version: Dated 2026-10-14; retrieved at commit 0dcc533cc9f8f76c90ecbfe944df816b6672328a (21 pages) Retrieved: 2026-11-06T09:52:00Z | not individually reviewed No individual claim review recorded author reported Source has a recorded evidence concern. Consult its source page before using the claim. Audit detailsField: attributes.origin Source artifact SHA-256: 0f10b390be399c341d1f7497c5a3ca0ac857ce78172c4732bdc86a9c66e46419 Hash scope: pdftotext -layout text layer, parsed by extract/parse.py with every row label and column count asserted. Inspected artifact |
|---|