English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Computational protein design, i.e. inferring novel and diverse proteinsequences consistent with a given structure, remains a major unsolvedchallenge. Recently, deep generative models that learn from sequences alone orfrom sequences and structures jointly have shown impressive performance on thistask. However, those models appear limited in terms of modeling structuralconstraints, capturing enough sequence diversity, or both. Here we considerthree recently proposed deep generative frameworks for protein design: (AR) thesequence-based autoregressive generative model, (GVP) the precisestructure-based graph neural network, and Fold2Seq that leverages a fuzzy andscale-free representation of a three-dimensional fold, while enforcingstructure-to-sequence (and vice versa) consistency. We benchmark these modelson the task of computational design of antibody sequences, which demanddesigning sequences with high diversity for functional implication. TheFold2Seq framework outperforms the two other baselines in terms of diversity ofthe designed sequences, while maintaining the typical fold.

Benchmarking deep generative models for diverse antibody sequence design