rewirebio.iobenchmarks
Method

ABYSSAL

ESM-2 network with an added light-attention mechanism, trained on the mega-scale stability dataset, which predicts the stability change on single mutation and satisfies the antisymmetry property.

Results are recorded on linked configurations and versions of this method. Their settings and evaluations are kept separate below.

1 configuration with results

Results on linked records

This method has no results of its own in this release. One linked record has results. They are not assigned to the underlying method: each record keeps its own settings and evaluations, and its scores are on its own page.

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 ABYSSAL. 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

Use this method

How it works, versions and access

Additional methodological details, versions and access terms have not yet been extracted into a structured profile for this method. 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.

1 evidence row matching the loaded filters

Claims, original sources and review scope · Release 2026-11-21-adc15377fab2
Property and statementOriginal source and locationReview and provenance
Description
ESM-2 network with an added light-attention mechanism, trained on the mega-scale stability dataset, which predicts the stability change on single mutation and satisfies the antisymmetry property.
Individual claims
Abstract: New mega dataset combined with deep neural network makes a progress in predicting impact of mutation on protein stability (ABYSSAL)

Original source ↗

qa-descriptions-20261011-d-source-abyssal: abstractText

Version: Europe PMC preprint record PPR/PPR590902 (title and abstract), bioRxiv 2023
Retrieved: 2026-10-11T01:30:45Z

source checked

["automated-source-review","ai-assisted-source-review"] · 2026-10-11

Audit details

Approved. Checked the 16 Pancotti-derived descriptions word for word against PMC8921618 section 2.2, ESM therm, RaSP and ELASPIC-2 against Chu 2024, ThermoMPNN against Dieckhaus 2024, and ABYSSAL and phyloP against their abstracts; all match. Licences on the new sources are supported by Europe PMC, the PMC XML and Crossref.

Field: description

Claim: metadata-correction-f6f3db11f0ddfbca9b96

Source artifact SHA-256: 6b8741e12f40c5074150beecba35f0000e6d39a88c270d9327adfbcc57cddabf

Hash scope: SHA-256 of one retrieval of a Europe PMC REST search response. These responses are not byte-stable (a later retrieval can differ in JSON key order); the title and abstractText fields are the evidence.

Inspected artifact

Sources and history

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

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

Technical metadata for this record is tracked under stable ID protein-stability-20261009-method-abyssal.

Source locator: Method lists and table row labels of the cited sources

Related records

How to cite

ABYSSAL. Rewire biological model benchmark database, release 2026-11-21-adc15377fab2. https://benchmarks.rewirebio.io/database/method/protein-stability-20261009-method-abyssal/

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