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

Protein secondary structure prediction (PSSP) is essential for proteinfunction analysis. However, for low homologous proteins, the PSSP suffers frominsufficient input features. In this paper, we explicitly import externalself-supervised knowledge for low homologous PSSP under the guidance ofresidue-wise profile fusion. In practice, we firstly demonstrate thesuperiority of profile over Position-Specific Scoring Matrix (PSSM) for lowhomologous PSSP. Based on this observation, we introduce the novelself-supervised BERT features as the pseudo profile, which implicitly involvesthe residue distribution in all native discovered sequences as thecomplementary features. Further-more, a novel residue-wise attention isspecially designed to adaptively fuse different features (i.e.,originallow-quality profile, BERT based pseudo profile), which not only takes fulladvantage of each feature but also avoids noise disturbance. Be-sides, thefeature consistency loss is proposed to accelerate the model learning frommultiple semantic levels. Extensive experiments confirm that our methodoutperforms state-of-the-arts (i.e.,4.7%forextremely low homologous cases onBC40 dataset).

Adaptive Residue-wise Profile Fusion for Low Homologous Protein SecondaryStructure Prediction Using External Knowledge