From Holo Pockets to Electron Density: GPT-style Drug Design with Density

Jiahao Chen, Letian Gao, Yanhao Zhu, Wenbiao Zhou, Bing Su, Zhi John Lu, Bo Huang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:14346-14366, 2026.

Abstract

Recent advances in generative modeling have enabled significant progress in structure-based drug design (SBDD). Existing methods typically condition molecule generation on empty binding pockets from holo complexes, overlooking informative components such as the filler (ligands and solvent). Here, we leverage low-resolution electron density (ED) derived from the filler as a physically grounded condition for de novo drug design. We consider two types of ED—calculated and cryo-EM/X-ray—obtainable from computational or experimental sources, supporting unified pre-training and experimental integration. Compared with rigid pocket representations, experimental ED naturally captures conformational flexibility and provides a more faithful description of the binding environment. Based on this, we introduce EDMolGPT, a decoder-only autoregressive framework that generates molecules from low-resolution ED point clouds. By grounding generation in physically meaningful density signals, EDMolGPT mitigates structural bias and produces molecules with 3D conformations. Evaluations on 101 biological targets verify the effectiveness. Our project page: https://jiahaochen1.github.io/EDMolGPT_Page/.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26ai, title = {From Holo Pockets to Electron Density: {GPT}-style Drug Design with Density}, author = {Chen, Jiahao and Gao, Letian and Zhu, Yanhao and Zhou, Wenbiao and Su, Bing and Lu, Zhi John and Huang, Bo}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {14346--14366}, 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/chen26ai/chen26ai.pdf}, url = {https://proceedings.mlr.press/v306/chen26ai.html}, abstract = {Recent advances in generative modeling have enabled significant progress in structure-based drug design (SBDD). Existing methods typically condition molecule generation on empty binding pockets from holo complexes, overlooking informative components such as the filler (ligands and solvent). Here, we leverage low-resolution electron density (ED) derived from the filler as a physically grounded condition for de novo drug design. We consider two types of ED—calculated and cryo-EM/X-ray—obtainable from computational or experimental sources, supporting unified pre-training and experimental integration. Compared with rigid pocket representations, experimental ED naturally captures conformational flexibility and provides a more faithful description of the binding environment. Based on this, we introduce EDMolGPT, a decoder-only autoregressive framework that generates molecules from low-resolution ED point clouds. By grounding generation in physically meaningful density signals, EDMolGPT mitigates structural bias and produces molecules with 3D conformations. Evaluations on 101 biological targets verify the effectiveness. Our project page: https://jiahaochen1.github.io/EDMolGPT_Page/.} }
Endnote
%0 Conference Paper %T From Holo Pockets to Electron Density: GPT-style Drug Design with Density %A Jiahao Chen %A Letian Gao %A Yanhao Zhu %A Wenbiao Zhou %A Bing Su %A Zhi John Lu %A Bo 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-chen26ai %I PMLR %P 14346--14366 %U https://proceedings.mlr.press/v306/chen26ai.html %V 306 %X Recent advances in generative modeling have enabled significant progress in structure-based drug design (SBDD). Existing methods typically condition molecule generation on empty binding pockets from holo complexes, overlooking informative components such as the filler (ligands and solvent). Here, we leverage low-resolution electron density (ED) derived from the filler as a physically grounded condition for de novo drug design. We consider two types of ED—calculated and cryo-EM/X-ray—obtainable from computational or experimental sources, supporting unified pre-training and experimental integration. Compared with rigid pocket representations, experimental ED naturally captures conformational flexibility and provides a more faithful description of the binding environment. Based on this, we introduce EDMolGPT, a decoder-only autoregressive framework that generates molecules from low-resolution ED point clouds. By grounding generation in physically meaningful density signals, EDMolGPT mitigates structural bias and produces molecules with 3D conformations. Evaluations on 101 biological targets verify the effectiveness. Our project page: https://jiahaochen1.github.io/EDMolGPT_Page/.
APA
Chen, J., Gao, L., Zhu, Y., Zhou, W., Su, B., Lu, Z.J. & Huang, B.. (2026). From Holo Pockets to Electron Density: GPT-style Drug Design with Density. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:14346-14366 Available from https://proceedings.mlr.press/v306/chen26ai.html.

Related Material