Generative Coarse-Graining of Molecular Conformations

Wujie Wang, Minkai Xu, Chen Cai, Benjamin K Miller, Tess Smidt, Yusu Wang, Jian Tang, Rafael Gomez-Bombarelli
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:23213-23236, 2022.

Abstract

Coarse-graining (CG) of molecular simulations simplifies the particle representation by grouping selected atoms into pseudo-beads and therefore drastically accelerates simulation. However, such CG procedure induces information losses, which makes accurate backmapping, i.e., restoring fine-grained (FG) coordinates from CG coordinates, a long-standing challenge. Inspired by the recent progress in generative models and equivariant networks, we propose a novel model that rigorously embeds the vital probabilistic nature and geometrical consistency requirements of the backmapping transformation. Our model encodes the FG uncertainties into an invariant latent space and decodes them back to FG geometries via equivariant convolutions. To standardize the evaluation of this domain, we further provide three comprehensive benchmarks based on molecular dynamics trajectories. Extensive experiments show that our approach always recovers more realistic structures and outperforms existing data-driven methods with a significant margin.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-wang22ag, title = {Generative Coarse-Graining of Molecular Conformations}, author = {Wang, Wujie and Xu, Minkai and Cai, Chen and Miller, Benjamin K and Smidt, Tess and Wang, Yusu and Tang, Jian and Gomez-Bombarelli, Rafael}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {23213--23236}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/wang22ag/wang22ag.pdf}, url = {https://proceedings.mlr.press/v162/wang22ag.html}, abstract = {Coarse-graining (CG) of molecular simulations simplifies the particle representation by grouping selected atoms into pseudo-beads and therefore drastically accelerates simulation. However, such CG procedure induces information losses, which makes accurate backmapping, i.e., restoring fine-grained (FG) coordinates from CG coordinates, a long-standing challenge. Inspired by the recent progress in generative models and equivariant networks, we propose a novel model that rigorously embeds the vital probabilistic nature and geometrical consistency requirements of the backmapping transformation. Our model encodes the FG uncertainties into an invariant latent space and decodes them back to FG geometries via equivariant convolutions. To standardize the evaluation of this domain, we further provide three comprehensive benchmarks based on molecular dynamics trajectories. Extensive experiments show that our approach always recovers more realistic structures and outperforms existing data-driven methods with a significant margin.} }
Endnote
%0 Conference Paper %T Generative Coarse-Graining of Molecular Conformations %A Wujie Wang %A Minkai Xu %A Chen Cai %A Benjamin K Miller %A Tess Smidt %A Yusu Wang %A Jian Tang %A Rafael Gomez-Bombarelli %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-wang22ag %I PMLR %P 23213--23236 %U https://proceedings.mlr.press/v162/wang22ag.html %V 162 %X Coarse-graining (CG) of molecular simulations simplifies the particle representation by grouping selected atoms into pseudo-beads and therefore drastically accelerates simulation. However, such CG procedure induces information losses, which makes accurate backmapping, i.e., restoring fine-grained (FG) coordinates from CG coordinates, a long-standing challenge. Inspired by the recent progress in generative models and equivariant networks, we propose a novel model that rigorously embeds the vital probabilistic nature and geometrical consistency requirements of the backmapping transformation. Our model encodes the FG uncertainties into an invariant latent space and decodes them back to FG geometries via equivariant convolutions. To standardize the evaluation of this domain, we further provide three comprehensive benchmarks based on molecular dynamics trajectories. Extensive experiments show that our approach always recovers more realistic structures and outperforms existing data-driven methods with a significant margin.
APA
Wang, W., Xu, M., Cai, C., Miller, B.K., Smidt, T., Wang, Y., Tang, J. & Gomez-Bombarelli, R.. (2022). Generative Coarse-Graining of Molecular Conformations. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:23213-23236 Available from https://proceedings.mlr.press/v162/wang22ag.html.

Related Material