Skip to content

data_availability.md

Latest commit

 

History

History
37 lines (24 loc) · 1.33 KB

data_availability.md

File metadata and controls

37 lines (24 loc) · 1.33 KB
Raw

Data Availability and Use Boundary

This page summarizes the AbBindGym repository data boundary.

Repository contents

This GitHub repository contains source code, synthetic examples, tests, public-safe aggregate result tables, and placeholder paths for external data.

It does not contain raw datasets, processed benchmark tables, sequence-level AbELA-Q records, per-record predictions, model checkpoints, prediction caches, generated candidates, or other materials governed by separate data-use terms.

Public source data

Raw public datasets should be downloaded from their original sources and cited according to the source publications and terms of use.

Processed benchmark data

Processed benchmark data used for the paper are provided through the Zenodo archive, not through GitHub:

https://doi.org/10.5281/zenodo.19911284

Controlled-access AbELA-Q

Sequence-level AbELA-Q data are controlled-access materials and are excluded from this repository. Access terms are described in the manuscript Data Availability statement.

Use boundary

The MIT License in LICENSE applies to AbBindGym source code only. It does not grant redistribution or reuse rights for third-party datasets, processed benchmark data, controlled-access AbELA-Q records, model weights, or materials covered by their own licences, terms of use, or data-use agreements.