bioRxivBioinformaticsPreprint
Foldy: a web application for interactive protein structure analysis
Jacob B. RobertsiD, Alberto A. NavaiD, Allison N. PearsoniD, Matthew R. InchaiD, Luis E. Valencia, Melody Ma, Abhay Rao, Jay D. KeaslingiD
- Citations
- 4
Abstract
Foldy is a cloud-based application that allows non-expert scientists to easily access and utilize advanced AI-based structural biology tools, including AlphaFold and DiffDock. Built on Kubernetes, it can be deployed by universities, departments, and labs without requiring hardware resources, but can also be configured to utilize available computers. Foldy enables scientists to predict the structure of proteins and complexes up to 3000 amino acids, visualize Pfam annotations, and dock ligands with AutoDock Vina and DiffDock. Our manuscript describes the user interface and deployment considerations of Foldy, as well as some of our applications. By democratizing access to sophisticated AI-based tools, Foldy can facilitate life science research and promote the wider adoption of structural bioinformatics tools. Our work demonstrates that even the most advanced tools can be made accessible to a broad audience through user-friendly platforms like Foldy, and we believe it will be a valuable resource for researchers across scientific disciplines. The public structures available on the Lawrence Berkeley Labs Foldy deployment can be viewed at https://foldy.lbl.gov. Author SummaryFoldy is a cloud-based application that enables scientists to use AI-based structural biology tools such as AlphaFold and DiffDock without software expertise. Built on Kubernetes, it can be set up by universities, departments, and labs with no need for hardware resources. Foldy can predict the structure of proteins and complexes up to 3000 amino acids, visualize Pfam annotations, and dock ligands with AutoDock Vina and DiffDock. Our public structures can be viewed at https://foldy.lbl.gov. Our manuscript highlights the user interface, deployment considerations, and product applications of Foldy. Its an accessible solution for researchers who are not software experts and can handle the traffic of thousands of users and hundreds of thousands of protein structures and docked ligands. This makes advanced AI-based tools more widely available, paving the way for accelerating life science research. By developing an easy-to-use platform, our work demonstrates that even the most sophisticated AI-based tools can be made accessible to a wide audience. Foldy enables more scientists to draw from the rapidly growing field of structural biology, making it a valuable tool for researchers across scientific disciplines. We look forward to its adoption by the scientific community.
- assay: protein structure
- research focus: software / tool
- research focus: benchmarking
- approach: experimental assay
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- ColabFold - Making protein folding accessible to all
Milot Mirdita, Konstantin Schütze, … Martin Steinegger·Nature MethodsPeer-reviewed·May 30, 2022
- FlashFold: a standalone command-line tool for accelerated protein structure and stoichiometry prediction
Chayan Kumar Saha, Mohammad Roghanian, … Lionel Guy·bioRxiv·Jan 23, 2025
- AF3 Inspector: In-Browser 3D Model Visualization and Confidence Analytics for AlphaFold 3
Xavier Gemis-Aebi, Franco Jiménez, … Luciano A. Abriata·bioRxiv·Sep 28, 2026
- ABCFold: easier running and comparison of AlphaFold 3, Boltz-1 and Chai-1
Luc G. Elliott, Adam J. Simpkin, Daniel John Rigden·Bioinformatics AdvancesPeer-reviewed·Dec 26, 2024
- Protein structure prediction and design for high-throughput computing
Vinay Saji Mathew, Gretta D. Kellogg, William Lai·bioRxiv·Jul 18, 2025
- PymolFold: A PyMOL Plugin for API-driven Structure Prediction and Quality Assessment
Yifan Deng, Jinyuan Sun·Journal of Chemical Information and ModelingPeer-reviewed·Jan 2, 2026