Protein Discovery with Discrete Walk-Jump Sampling
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866915196612640768 |
|---|---|
| author | Frey, Nathan C. Berenberg, Daniel Zadorozhny, Karina Kleinhenz, Joseph Lafrance-Vanasse, Julien Hotzel, Isidro Wu, Yan Ra, Stephen Bonneau, Richard Cho, Kyunghyun Loukas, Andreas Gligorijevic, Vladimir Saremi, Saeed |
| author_facet | Frey, Nathan C. Berenberg, Daniel Zadorozhny, Karina Kleinhenz, Joseph Lafrance-Vanasse, Julien Hotzel, Isidro Wu, Yan Ra, Stephen Bonneau, Richard Cho, Kyunghyun Loukas, Andreas Gligorijevic, Vladimir Saremi, Saeed |
| contents | We resolve difficulties in training and sampling from a discrete generative model by learning a smoothed energy function, sampling from the smoothed data manifold with Langevin Markov chain Monte Carlo (MCMC), and projecting back to the true data manifold with one-step denoising. Our Discrete Walk-Jump Sampling formalism combines the contrastive divergence training of an energy-based model and improved sample quality of a score-based model, while simplifying training and sampling by requiring only a single noise level. We evaluate the robustness of our approach on generative modeling of antibody proteins and introduce the distributional conformity score to benchmark protein generative models. By optimizing and sampling from our models for the proposed distributional conformity score, 97-100% of generated samples are successfully expressed and purified and 70% of functional designs show equal or improved binding affinity compared to known functional antibodies on the first attempt in a single round of laboratory experiments. We also report the first demonstration of long-run fast-mixing MCMC chains where diverse antibody protein classes are visited in a single MCMC chain. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_12360 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Protein Discovery with Discrete Walk-Jump Sampling Frey, Nathan C. Berenberg, Daniel Zadorozhny, Karina Kleinhenz, Joseph Lafrance-Vanasse, Julien Hotzel, Isidro Wu, Yan Ra, Stephen Bonneau, Richard Cho, Kyunghyun Loukas, Andreas Gligorijevic, Vladimir Saremi, Saeed Biomolecules Machine Learning We resolve difficulties in training and sampling from a discrete generative model by learning a smoothed energy function, sampling from the smoothed data manifold with Langevin Markov chain Monte Carlo (MCMC), and projecting back to the true data manifold with one-step denoising. Our Discrete Walk-Jump Sampling formalism combines the contrastive divergence training of an energy-based model and improved sample quality of a score-based model, while simplifying training and sampling by requiring only a single noise level. We evaluate the robustness of our approach on generative modeling of antibody proteins and introduce the distributional conformity score to benchmark protein generative models. By optimizing and sampling from our models for the proposed distributional conformity score, 97-100% of generated samples are successfully expressed and purified and 70% of functional designs show equal or improved binding affinity compared to known functional antibodies on the first attempt in a single round of laboratory experiments. We also report the first demonstration of long-run fast-mixing MCMC chains where diverse antibody protein classes are visited in a single MCMC chain. |
| title | Protein Discovery with Discrete Walk-Jump Sampling |
| topic | Biomolecules Machine Learning |
| url | https://arxiv.org/abs/2306.12360 |