RecCrysFormer: Refined Protein Structural Prediction from 3D Patterson Maps via Recycling Training Runs

Tom Pan, Evan Dramko, Mitchell D. Miller, George N Phillips Jr., Anastasios Kyrillidis
Conference on Parsimony and Learning, PMLR 280:897-912, 2025.

Abstract

Determining protein structures at an atomic level remains a significant challenge in structural biology. We introduce $\texttt{RecCrysFormer}$, a hybrid model that exploits the strengths of transformers with the aim of integrating experimental and ML approaches to protein structure determination from crystallographic data. $\texttt{RecCrysFormer}$ leverages Patterson maps and incorporates known standardized partial structures of amino acid residues to directly predict electron density maps, which are essential for constructing detailed atomic models through crystallographic refinement processes. $\texttt{RecCrysFormer}$ benefits from a “recycling” training regimen that iteratively incorporates results from crystallographic refinements and previous training runs as additional inputs in the form of template maps. Using a preliminary dataset of synthetic peptide fragments based on Protein Data Bank, $\texttt{RecCrysFormer}$ achieves good accuracy in structural predictions and shows robustness against variations in crystal parameters, such as unit cell dimensions and angles.

Cite this Paper

BibTeX
@InProceedings{pmlr-v280-pan25a, title = {RecCrysFormer: Refined Protein Structural Prediction from 3D Patterson Maps via Recycling Training Runs}, author = {Pan, Tom and Dramko, Evan and Miller, Mitchell D. and Jr., George N Phillips and Kyrillidis, Anastasios}, booktitle = {Conference on Parsimony and Learning}, pages = {897--912}, year = {2025}, editor = {Chen, Beidi and Liu, Shijia and Pilanci, Mert and Su, Weijie and Sulam, Jeremias and Wang, Yuxiang and Zhu, Zhihui}, volume = {280}, series = {Proceedings of Machine Learning Research}, month = {24--27 Mar}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v280/main/assets/pan25a/pan25a.pdf}, url = {https://proceedings.mlr.press/v280/pan25a.html}, abstract = {Determining protein structures at an atomic level remains a significant challenge in structural biology. We introduce $\texttt{RecCrysFormer}$, a hybrid model that exploits the strengths of transformers with the aim of integrating experimental and ML approaches to protein structure determination from crystallographic data. $\texttt{RecCrysFormer}$ leverages Patterson maps and incorporates known standardized partial structures of amino acid residues to directly predict electron density maps, which are essential for constructing detailed atomic models through crystallographic refinement processes. $\texttt{RecCrysFormer}$ benefits from a “recycling” training regimen that iteratively incorporates results from crystallographic refinements and previous training runs as additional inputs in the form of template maps. Using a preliminary dataset of synthetic peptide fragments based on Protein Data Bank, $\texttt{RecCrysFormer}$ achieves good accuracy in structural predictions and shows robustness against variations in crystal parameters, such as unit cell dimensions and angles.} }
Endnote
%0 Conference Paper %T RecCrysFormer: Refined Protein Structural Prediction from 3D Patterson Maps via Recycling Training Runs %A Tom Pan %A Evan Dramko %A Mitchell D. Miller %A George N Phillips Jr. %A Anastasios Kyrillidis %B Conference on Parsimony and Learning %C Proceedings of Machine Learning Research %D 2025 %E Beidi Chen %E Shijia Liu %E Mert Pilanci %E Weijie Su %E Jeremias Sulam %E Yuxiang Wang %E Zhihui Zhu %F pmlr-v280-pan25a %I PMLR %P 897--912 %U https://proceedings.mlr.press/v280/pan25a.html %V 280 %X Determining protein structures at an atomic level remains a significant challenge in structural biology. We introduce $\texttt{RecCrysFormer}$, a hybrid model that exploits the strengths of transformers with the aim of integrating experimental and ML approaches to protein structure determination from crystallographic data. $\texttt{RecCrysFormer}$ leverages Patterson maps and incorporates known standardized partial structures of amino acid residues to directly predict electron density maps, which are essential for constructing detailed atomic models through crystallographic refinement processes. $\texttt{RecCrysFormer}$ benefits from a “recycling” training regimen that iteratively incorporates results from crystallographic refinements and previous training runs as additional inputs in the form of template maps. Using a preliminary dataset of synthetic peptide fragments based on Protein Data Bank, $\texttt{RecCrysFormer}$ achieves good accuracy in structural predictions and shows robustness against variations in crystal parameters, such as unit cell dimensions and angles.
APA
Pan, T., Dramko, E., Miller, M.D., Jr., G.N.P. & Kyrillidis, A.. (2025). RecCrysFormer: Refined Protein Structural Prediction from 3D Patterson Maps via Recycling Training Runs. Conference on Parsimony and Learning, in Proceedings of Machine Learning Research 280:897-912 Available from https://proceedings.mlr.press/v280/pan25a.html.

Related Material