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-protein interactions (PPIs) are essentials for many biologicalprocesses where two or more proteins physically bind together to achieve theirfunctions. Modeling PPIs is useful for many biomedical applications, such asvaccine design, antibody therapeutics, and peptide drug discovery. Pre-traininga protein model to learn effective representation is critical for PPIs. Mostpre-training models for PPIs are sequence-based, which naively adopt thelanguage models used in natural language processing to amino acid sequences.More advanced works utilize the structure-aware pre-training technique, takingadvantage of the contact maps of known protein structures. However, neithersequences nor contact maps can fully characterize structures and functions ofthe proteins, which are closely related to the PPI problem. Inspired by thisinsight, we propose a multimodal protein pre-training model with threemodalities: sequence, structure, and function (S2F). Notably, instead of usingcontact maps to learn the amino acid-level rigid structures, we encode thestructure feature with the topology complex of point clouds of heavy atoms. Itallows our model to learn structural information about not only the backbonesbut also the side chains. Moreover, our model incorporates the knowledge fromthe functional description of proteins extracted from literature or manualannotations. Our experiments show that the S2F learns protein embeddings thatachieve good performances on a variety of PPIs tasks, including cross-speciesPPI, antibody-antigen affinity prediction, antibody neutralization predictionfor SARS-CoV-2, and mutation-driven binding affinity change prediction.

Multimodal Pre-Training Model for Sequence-based Prediction of Protein-Protein Interaction