# Protein Fold Classification at Scale: Benchmarking and Pretraining

arXiv: https://arxiv.org/abs/2605.18552 · HTML: https://mlsanity.com/paper/2605.18552/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 8 of its 48 references that are in the map.
- First posted on arXiv 18 May 2026.

## Builds on (most cited first, 8 of 8)

- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges](https://mlsanity.com/paper/2104.13478/) (2021)
- [MACE: Higher Order Equivariant Message Passing Neural Networks for Fast and Accurate Force Fields](https://mlsanity.com/paper/2206.07697/) (2022)
- [Evaluating Protein Transfer Learning with TAPE](https://mlsanity.com/paper/1906.08230/) (2019)
- [Protein Representation Learning by Geometric Structure Pretraining](https://mlsanity.com/paper/2203.06125/) (2022)
- [DeepSF: deep convolutional neural network for mapping protein sequences to folds](https://mlsanity.com/paper/1706.01010/) (2017)
- [Endowing Protein Language Models with Structural Knowledge](https://mlsanity.com/paper/2401.14819/) (2024)

## Built on by (most cited first, 2 of 2)

- [Traditional machine learning vs. deep learning from dynamic graph representations of proteins' 3D folds in the task of protein structure classification](https://mlsanity.com/paper/2605.29228/) (2026)
- [Equivariance Breaks the Learning Rate](https://mlsanity.com/paper/2609.08381/) (2026)

## Abstract (arXiv)

Classifying protein topology is essential for deciphering biological function, but progress is held back by the lack of large-scale benchmarks that avoid duplicates and by models that do not scale well. We introduce TEDBench, a large-scale, non-redundant benchmark for protein fold classification constructed from the Encyclopedia of Domains (TED) and Foldseek-clustered AlphaFold structures. We show that on TEDBench, current protein representation learning methods either require very large models or fail to deliver strong performance. To address this challenge, we propose Masked Invariant Autoencoders (MiAE), a self-supervised framework for protein structure representation learning. MiAE uses an extremely high masking ratio of up to 90% with an SE(3)-invariant encoder and a lightweight decoder that reconstructs backbone coordinates from the latent representation and mask tokens. MiAE scales well and outperforms supervised counterparts and state-of-the-art baselines on TEDBench, establishing a strong recipe for protein fold classification. To test transfer beyond AlphaFold structures, we further benchmark on a curated dataset from experimental structures of CATH v4.4. TEDBench is available at https://github.com/BorgwardtLab/TEDBench.