Trending

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and Machine Learning and more.Subscribe

Artificial Intelligence

Diffusion-based generative model for protein backbone structure

Protein structure generation via folding diffusion

We present a new diffusion-based model that unconditionally designs protein backbone structures via a procedure that mirrors the native folding process.

We describe protein backbone structure as a series of consecutive angles capturing the relative orientation of the amino acid residues, and generate new structures by denoising from a random, unfolded state towards a stable folded structure.

The inherent shift and rotational invariance of this representation crucially alleviates the need for complex equivariant networks.

Authors

Kevin E. Wu, Kevin K. Yang, Rianne van den Berg, James Y. Zou, Alex X. Lu, Ava P. Amini

Related Topics

Neural networks

Diffusion-based generative model

Denoising diffusion probabilistic model

Complex equivariant networks

Read the Paper

Read Paper

Trending

Bookmarks

◐ Recommended Follows

Founder 42papers.com. Also bui...

More

◐ Latest Activity

More

Diffusion-based generative model for protein backbone structure - 42Papers

Read Related Papers

Hidden-Fold Networks: Random Recurrent Residuals Using Sparse Supermasks

A Unified Approach to Variational Autoencoders and Stochastic Normalizing Flows via Markov Chains

Multiscale pattern formation using mass-conserving reaction-diffusion systems

Efficient Decompositional Rule Extraction for Deep Neural Networks

A gentle introduction to the non-equilibrium physics of trajectories: Theory, algorithms, and biomolecular applications

Deep generative estimation of free energy in lattice field theories

Lossless Compression with Probabilistic Circuits

SOM-VAE: Interpretable Discrete Representation Learning on Time Series

Smoothing the Generative Latent Space with Mixup-based Distance Learning

Molecular Dynamics Analysis of Biomolecular Systems Including Nucleic Acids