AlphaFold Database introduces sequence-based search and structure cluster members

DA
David Armstrong
Thu, Oct 26, 2023 3:56 PM

**

Dear pdb-l

We are excited to announce two significant functionalities added to the
AlphaFold Protein Structure Database (AlphaFold DB).

The database has expanded since its inception in 2021 to cover over 214
million predicted structures from Google Deepmind’s AlphaFold 2 system.
We now introduce sequence similarity-based search and enhanced
prediction pages by displaying structurally similar predictions.

The sequence-similarity search is powered by the Basic Local Alignment
Search Tool (BLAST
https://blast.ncbi.nlm.nih.gov/doc/blast-help/downloadblastdata.html)
which enables finding relevant predicted structures based on a protein
sequence. Sequence search results have been integrated to our current
search results pages, highlighting reviewed status and reference proteome.

In a collaboration with the Steinegger lab and to make the navigation of
the structural landscape easier, we have incorporated the clustering
algorithm, Foldseek Cluster
https://www.nature.com/articles/s41586-023-06510-w. For a streamlined
experience, we have added the new structure similarity cluster feature
that has been added directly to the prediction pages.

We believe that these new features will make AlphaFold DB even more
useful and accessible to the scientific community. To learn more, please
visit the AlphaFold DB website at https://alphafold.ebi.ac.uk/.

For more information about the new features, visit:
https://www.ebi.ac.uk/about/news/updates-from-data-resources/alphafold-database-update-sequence-based-search/
https://www.ebi.ac.uk/about/news/updates-from-data-resources/alphafold-database-update-sequence-based-search/

Kind regards, David Armstrong

--
David Armstrong
Outreach and Training Lead
PDBe
European Bioinformatics Institute (EMBL-EBI)
European Molecular Biology Laboratory
Wellcome Trust Genome Campus
Hinxton
Cambridge CB10 1SD UK

** *Dear pdb-l* * We are excited to announce two significant functionalities added to the AlphaFold Protein Structure Database (AlphaFold DB). The database has expanded since its inception in 2021 to cover over 214 million predicted structures from Google Deepmind’s AlphaFold 2 system. We now introduce sequence similarity-based search and enhanced prediction pages by displaying structurally similar predictions. The sequence-similarity search is powered by the Basic Local Alignment Search Tool (BLAST <https://blast.ncbi.nlm.nih.gov/doc/blast-help/downloadblastdata.html>) which enables finding relevant predicted structures based on a protein sequence. Sequence search results have been integrated to our current search results pages, highlighting reviewed status and reference proteome. In a collaboration with the Steinegger lab and to make the navigation of the structural landscape easier, we have incorporated the clustering algorithm, Foldseek Cluster <https://www.nature.com/articles/s41586-023-06510-w>. For a streamlined experience, we have added the new structure similarity cluster feature that has been added directly to the prediction pages. We believe that these new features will make AlphaFold DB even more useful and accessible to the scientific community. To learn more, please visit the AlphaFold DB website at https://alphafold.ebi.ac.uk/. For more information about the new features, visit: https://www.ebi.ac.uk/about/news/updates-from-data-resources/alphafold-database-update-sequence-based-search/ <https://www.ebi.ac.uk/about/news/updates-from-data-resources/alphafold-database-update-sequence-based-search/> Kind regards, David Armstrong * -- David Armstrong Outreach and Training Lead PDBe European Bioinformatics Institute (EMBL-EBI) European Molecular Biology Laboratory Wellcome Trust Genome Campus Hinxton Cambridge CB10 1SD UK