Autoregressive fragment-based diffusion for pocket-aware ligand design
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_ | 1866914637221462016 |
|---|---|
| author | Ghorbani, Mahdi Gendelev, Leo Beroza, Paul Keiser, Michael J. |
| author_facet | Ghorbani, Mahdi Gendelev, Leo Beroza, Paul Keiser, Michael J. |
| contents | In this work, we introduce AutoFragDiff, a fragment-based autoregressive diffusion model for generating 3D molecular structures conditioned on target protein structures. We employ geometric vector perceptrons to predict atom types and spatial coordinates of new molecular fragments conditioned on molecular scaffolds and protein pockets. Our approach improves the local geometry of the resulting 3D molecules while maintaining high predicted binding affinity to protein targets. The model can also perform scaffold extension from user-provided starting molecular scaffold. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_05370 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Autoregressive fragment-based diffusion for pocket-aware ligand design Ghorbani, Mahdi Gendelev, Leo Beroza, Paul Keiser, Michael J. Biomolecules Artificial Intelligence Machine Learning Chemical Physics Quantitative Methods J.2; I.2.6; I.6.3 In this work, we introduce AutoFragDiff, a fragment-based autoregressive diffusion model for generating 3D molecular structures conditioned on target protein structures. We employ geometric vector perceptrons to predict atom types and spatial coordinates of new molecular fragments conditioned on molecular scaffolds and protein pockets. Our approach improves the local geometry of the resulting 3D molecules while maintaining high predicted binding affinity to protein targets. The model can also perform scaffold extension from user-provided starting molecular scaffold. |
| title | Autoregressive fragment-based diffusion for pocket-aware ligand design |
| topic | Biomolecules Artificial Intelligence Machine Learning Chemical Physics Quantitative Methods J.2; I.2.6; I.6.3 |
| url | https://arxiv.org/abs/2401.05370 |