Cargando…
Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention
The established approach to unsupervised protein contact prediction estimates coevolving positions using undirected graphical models. This approach trains a Potts model on a Multiple Sequence Alignment. Increasingly large Transformers are being pretrained on unlabeled, unaligned protein sequence dat...
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Online Artículo Texto |
| Lenguaje: | English |
| Publicado: |
2022
|
| Materias: | |
| Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8752338/ https://www.ncbi.nlm.nih.gov/pubmed/34890134 |