# Learning Geometrically Disentangled Representations of Protein Folding Simulations

arXiv: https://arxiv.org/abs/2205.10423 · HTML: https://mlsanity.com/paper/2205.10423/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 33 references that are in the map.
- First posted on arXiv 20 May 2022.

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Graph Attention Networks](https://mlsanity.com/paper/1710.10903/) (2017)
- [Fast R-CNN](https://mlsanity.com/paper/1504.08083/) (2015)
- [Learning from Protein Structure with Geometric Vector Perceptrons](https://mlsanity.com/paper/2009.01411/) (2020)
- [Symplectic Recurrent Neural Networks](https://mlsanity.com/paper/1909.13334/) (2019)
- [VAMPnets: Deep learning of molecular kinetics](https://mlsanity.com/paper/1710.06012/) (2017)
- [Differentiable Molecular Simulations for Control and Learning](https://mlsanity.com/paper/2003.00868/) (2020)
- [Molecular enhanced sampling with autoencoders: On-the-fly collective variable discovery and accelerated free energy landscape exploration](https://mlsanity.com/paper/1801.00203/) (2017)
- [DeepDriveMD: Deep-Learning Driven Adaptive Molecular Simulations for Protein Folding](https://mlsanity.com/paper/1909.07817/) (2019)
- [Nonlinear Discovery of Slow Molecular Modes using State-Free Reversible VAMPnets](https://mlsanity.com/paper/1902.03336/) (2019)
- [Interpretable Embeddings From Molecular Simulations Using Gaussian Mixture Variational Autoencoders](https://mlsanity.com/paper/1912.12175/) (2019)
- [Disentangling Content and Style via Unsupervised Geometry Distillation](https://mlsanity.com/paper/1905.04538/) (2019)
- [Generating Tertiary Protein Structures via an Interpretative Variational Autoencoder](https://mlsanity.com/paper/2004.07119/) (2020)
- [Unsupervised Geometric Disentanglement for Surfaces via CFAN-VAE](https://mlsanity.com/paper/2005.11622/) (2020)
- [Intrinsic-Extrinsic Convolution and Pooling for Learning on 3D Protein Structures](https://mlsanity.com/paper/2007.06252/) (2020)
- [Learning protein conformational space by enforcing physics with convolutions and latent interpolations](https://mlsanity.com/paper/1910.04543/) (2019)

## Abstract (arXiv)

Massive molecular simulations of drug-target proteins have been used as a tool to understand disease mechanism and develop therapeutics. This work focuses on learning a generative neural network on a structural ensemble of a drug-target protein, e.g. SARS-CoV-2 Spike protein, obtained from computationally expensive molecular simulations. Model tasks involve characterizing the distinct structural fluctuations of the protein bound to various drug molecules, as well as efficient generation of protein conformations that can serve as an complement of a molecular simulation engine. Specifically, we present a geometric autoencoder framework to learn separate latent space encodings of the intrinsic and extrinsic geometries of the protein structure. For this purpose, the proposed Protein Geometric AutoEncoder (ProGAE) model is trained on the protein contact map and the orientation of the backbone bonds of the protein. Using ProGAE latent embeddings, we reconstruct and generate the conformational ensemble of a protein at or near the experimental resolution, while gaining better interpretability and controllability in term of protein structure generation from the learned latent space. Additionally, ProGAE models are transferable to a different state of the same protein or to a new protein of different size, where only the dense layer decoding from the latent representation needs to be retrained. Results show that our geometric learning-based method enjoys both accuracy and efficiency for generating complex structural variations, charting the path toward scalable and improved approaches for analyzing and enhancing high-cost simulations of drug-target proteins.