New in Expasy: TEA (and STEAM)

Posted 17 June 2026
New in Expasy: TEA (and STEAM)

Interested in carrying out highly sensitive and efficient large-scale protein homology searches? STEAM, Search with TEA against Many, performs a fast search against large datasets of proteins translated to TEA (The Embedded Alphabet).

The preprint Rewriting protein alphabets with language models introduces a novel approach using contrastive learning to convert protein language model embeddings into a new 20-letter alphabet, TEA, enabling highly efficient large-scale protein homology searches.

Share this article

Latest News

Updates to ViralZone

Posted 19 November 2026
Updates to ViralZone

ViralZone, a resource for all viral genus and families, has updated the taxonomy for monodnaviria, added information on polyprotein cleavage prediction and detailed virus fact sheets for Ebolavirus and Andes virus. [Read more]

Updates to UniLectin

Posted 04 November 2026
Updates to UniLectin

The UniLectin3D and PathoLec datasets have been updated. UniLectin3D has been enriched with 27 new structures, which has led to changes in lectin classification. PathoLec coverage of pathogen agents, host organisms, affected tissues, diseases, and glycan binding ligands has been broadened with the addition of 195 lectin entries. Fold, class and family annotations were reviewed and corrected across the full dataset.

Participate in the EPIC challenge

EPIC (Eukaryotic Promoter and transcription Initiation prediction Challenge) aims to establish a rigorous benchmark for promoter and transcription initiation prediction in understudied species, advance sequence-based genomic modelling, and improve our understanding of how regulatory information is encoded in eukaryotic genomes. The closing date is Dec 31, 2026 so you can still join and win!