SLAE: Strictly Local All-atom Environment for Protein Representation

Yilin Chen, Tianyu Lu, Cizhang Zhao, Hannah Wayment-Steele, Po-Ssu Huang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:15919-15944, 2026.

Abstract

Building physically grounded protein representations is central to computational biology, yet most existing approaches rely on sequence-pretrained language models or backbone-only graphs that overlook side-chain geometry and chemical detail. We present SLAE, a unified all-atom framework for learning protein representations from each residue’s local atomic neighborhood using only atom types and interatomic geometries. To encourage expressive feature extraction, we introduce a novel multi-task autoencoder objective that combines coordinate reconstruction, sequence recovery, and energy regression. SLAE reconstructs allatom structures with high fidelity from latent residue environments and achieves state-of-the-art performance across diverse downstream tasks via transfer learning. SLAE’s latent space is chemically informative and environmentally sensitive, enabling quantitative assessment of structural qualities and smooth interpolation between conformations at all-atom resolution.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26ct, title = {{SLAE}: Strictly Local All-atom Environment for Protein Representation}, author = {Chen, Yilin and Lu, Tianyu and Zhao, Cizhang and Wayment-Steele, Hannah and Huang, Po-Ssu}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {15919--15944}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/chen26ct/chen26ct.pdf}, url = {https://proceedings.mlr.press/v306/chen26ct.html}, abstract = {Building physically grounded protein representations is central to computational biology, yet most existing approaches rely on sequence-pretrained language models or backbone-only graphs that overlook side-chain geometry and chemical detail. We present SLAE, a unified all-atom framework for learning protein representations from each residue’s local atomic neighborhood using only atom types and interatomic geometries. To encourage expressive feature extraction, we introduce a novel multi-task autoencoder objective that combines coordinate reconstruction, sequence recovery, and energy regression. SLAE reconstructs allatom structures with high fidelity from latent residue environments and achieves state-of-the-art performance across diverse downstream tasks via transfer learning. SLAE’s latent space is chemically informative and environmentally sensitive, enabling quantitative assessment of structural qualities and smooth interpolation between conformations at all-atom resolution.} }
Endnote
%0 Conference Paper %T SLAE: Strictly Local All-atom Environment for Protein Representation %A Yilin Chen %A Tianyu Lu %A Cizhang Zhao %A Hannah Wayment-Steele %A Po-Ssu Huang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-chen26ct %I PMLR %P 15919--15944 %U https://proceedings.mlr.press/v306/chen26ct.html %V 306 %X Building physically grounded protein representations is central to computational biology, yet most existing approaches rely on sequence-pretrained language models or backbone-only graphs that overlook side-chain geometry and chemical detail. We present SLAE, a unified all-atom framework for learning protein representations from each residue’s local atomic neighborhood using only atom types and interatomic geometries. To encourage expressive feature extraction, we introduce a novel multi-task autoencoder objective that combines coordinate reconstruction, sequence recovery, and energy regression. SLAE reconstructs allatom structures with high fidelity from latent residue environments and achieves state-of-the-art performance across diverse downstream tasks via transfer learning. SLAE’s latent space is chemically informative and environmentally sensitive, enabling quantitative assessment of structural qualities and smooth interpolation between conformations at all-atom resolution.
APA
Chen, Y., Lu, T., Zhao, C., Wayment-Steele, H. & Huang, P.. (2026). SLAE: Strictly Local All-atom Environment for Protein Representation. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:15919-15944 Available from https://proceedings.mlr.press/v306/chen26ct.html.

Related Material