GRIMM: Genetic stRatification for Inference in Molecular Modeling
Ashley Babjac, Adrienne Hoarfrost · 2026
Read GRIMM: Genetic stRatification for Inference in Molecular Modeling by Ashley Babjac, Adrienne Hoarfrost (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2602-16504.txt.
Description
Ashley Babjac and Adrienne Hoarfrost introduce GRIMM, a benchmark for enzyme function prediction built on genetic stratification. Most biological sequences encode unknown functions and look little like characterized proteins. So predicting enzyme function from sequence is hard, and models trained on public data can overstate performance and fail to generalize.
GRIMM clusters sequences by similarity and assigns whole clusters to training, validation, or test sets only, so no cluster appears in more than one split. It offers a closed-set test with the training label mix and an open-set test with novel labels, as a realistic proxy for discovering new enzyme functions. The approach is lightweight, needing only clustering and label annotations, and it generalizes to other sequence classification tasks.
Catalog record
| CREATOR | Ashley Babjac, Adrienne Hoarfrost |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-16504