Collaborators
The AlphaFold Protein Structure Database works with research groups and scientific consortia that contribute specialist expertise, high-quality structure predictions, annotations and tools.
AllTheBacteria project
AllTheBacteria is a collaborative, open-access project that assembles and annotates bacterial and archaeal genomes from public sequencing data.
Wheeler Lab, University of Edinburgh
The Wheeler Lab studies the cell biology of Leishmania, Trypanosoma and related parasites. The lab contributed protein structure predictions generated using enhanced multiple sequence alignments, improving predictions for proteins from these evolutionarily under-represented organisms.
Steinegger Lab, Seoul National University
The Steinegger Lab develops scalable, open-source methods for analysing biological sequences and structures. Its contributions to the AFDB include Foldseek, which enables rapid structure-similarity searches across the database; AFDB Clusters, which groups predicted structures into structural clusters to reveal distant relationships; and the AFDB50 collection, which groups proteins at 50% sequence identity to support efficient exploration of structural space.
Mirdita Lab, Sungkyunkwan University School of Medicine
Milot Mirdita leads the Laboratory for Computational Biology and Molecular Machine Learning. The group develops scalable, open bioinformatics infrastructure for sequence analysis, clustering and structure prediction. Mirdita co-developed ColabFold, which is used by several community datasets represented in AFDB, and collaborated on the proteome-scale homodimer and heterodimer predictions.
NVIDIA
NVIDIA develops accelerated computing methods for scientific research. Its researchers collaborated with EMBL-EBI, Google DeepMind and academic partners to adapt and run structure-prediction workflows at the scale required for AFDB protein-complex predictions, including homodimers and heterodimers.
Atkinson Lab, Lund University
The Atkinson Lab uses computational and experimental approaches to study protein function and evolution, with a particular interest in microbial attack and defence systems.
CATH and Jones groups, University College London
The CATH group led by Christine Orengo and the group led by David Jones jointly developed TED. TED identifies structural domains in AlphaFold models and maps them to known CATH classifications where possible, providing the domain annotations displayed on AFDB entry pages.
Grove Lab, MRC–University of Glasgow Centre for Virus Research
The Grove Lab combines at-scale protein structure prediction and classical experimental virology to investigate molecular mechanisms in viruses. In particular, the group focuses on viral entry glycoproteins and their interactions with the host. Along with colleagues at the Centre for Virus Research, The Grove Lab contributed the Viro3D dataset to the AFDB, and collaborated on proteome-level complex prediction for priority viral families.
Philippe Le Mercier and the ViralZone team, SIB Swiss Institute of Bioinformatics
Philippe Le Mercier leads the ViralZone team within SIB's Swiss-Prot group. ViralZone develops and curates comprehensive knowledge resources for virology, including virion structure and taxonomy, genome organisation, viral enzymes, and the mapping of polyprotein cleavage sites, all linked to viral protein records in UniProtKB.
