ColabFold: making protein folding accessible to all
By Milot Mirdita, Sergey Ovchinnikov, Martin Steinegger. 2022
7,715 citations recorded by the source corpus.
ColabFold offers accelerated prediction of protein structures and complexes by combining the fast homology search of MMseqs2 with AlphaFold2 or RoseTTAFold. ColabFold’s 40−60-fold faster search and optimized model utilization enables prediction of close to 1,000 structures per day on a server with one graphics processing unit. Coupled with Google Colaboratory, ColabFold becomes a free and accessible platform for protein folding. ColabFold is open-source software available at https://github.com/sokrypton/ColabFold and its novel environmental databases are available at https://colabfold.mmseqs.com. ColabFold is a free and accessible platform for protein folding that provides accelerated prediction of protein structures and complexes using AlphaFold2 or RoseTTAFold.
Subfields: Molecular Biology, Computational Biology & Bioinformatics, AI for Physical Sciences
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Molecular Biology
- Björn Wallner
- David T. Jones
- Arne Elofsson
- Yang Zhang
- David Baker
- Jinbo Xu
- Alexandre M. J. J. Bonvin
- Sergey Ovchinnikov
In Computational Biology & Bioinformatics
- David T. Jones
- Björn Wallner
- Joe G. Greener
- Arne Elofsson
- Jinbo Xu
- Bonnie Berger
- Kevin Kaichuang Yang
- Sergey Ovchinnikov
In AI for Physical Sciences
- Hassane S. Mchaourab
- Enrique Marcos
- Lim Heo
- Michael Feig
- Richard A. Stein
- Daniel Peñas-Utrilla
- Björn Wallner
- Jens Meiler
Citations in the corpus
- Predicting multiple conformations via sequence clustering and AlphaFold2 (2023, positive): “MSAs were generated using the MMseqs2-based 60 routine implemented in ColabFold 34 .”
- AFsample: improving multimer prediction with AlphaFold using massive sampling (2023, background): “In such cases, increasing the number of times the prediction is recycled in the network can improve performance (Mirdita et al. 2022).”
- In silico protein interaction screening uncovers DONSON’s role in replication initiation (2023, methodology): “5 (62) running on a Linux server equipped with 40GB NVIDIA A100 GPUs.”
- Programmable protein delivery with a bacterial contractile injection system (2023, positive): “To predict the structure of novel PVC tail fibre designs, we leveraged ColabFold, a Google Colab-based implementation of AlphaFold2 35–37 .”
- High-resolution de novo structure prediction from primary sequence (2022, methodology): “Sequential masking, where we mask either the first half or the second half of the sequence, akin to Prefix Language Modeling ( 4, 40 ) Moreover, we assign different weights for these loss terms.”
- Sampling alternative conformational states of transporters and receptors with AlphaFold2 (2022, background): “0.1 and a modified version of ColabFold ( Mirdita et al., 2021 ) that is available at https://github.com/delalamo/af2_conformations, ( Del Alamo, 2021b copy archived at swh:1:rev:d60db86886186e80622deaa91045caccaf4103d3).”
- Multi-State Modeling of G-protein Coupled Receptors at Experimental Accuracy (2021, methodology): “Various conformations were sampled by following the protocol described in Del Alamo et al.36 For a target GPCR, a MSA was generated by the ColabFold MMseqs2 API.41 From the original MSA, shallow MSAs were subsampled by modifying the AF2 network parameters, max_msa_clusters and max_extra_msa.”
- Protein complex prediction with AlphaFold-Multimer (2021, background): “AlphaFold-Gap (ColabFold [5]), version from 2021-08-16, is a published system that runs AlphaFold with a gap between residue indices between chains, uses MMSeqs2 for genetics, includes MSA pairing and does not include templates.”
- DeepTracer ID: De Novo Protein Identification from Cryo-EM Maps (2022, methodology): “For the other organisms, the user can either generate their AlphaFold2 library locally or use an online service such as ColabFold 26 .”
- Structural Predictions of the SNX-RGS Proteins Suggest They Belong to a New Class of Lipid Transfer Proteins (2021, positive): “We therefore also performed predictions of the full SNX25 sequence using the online ColabFold pipeline (Mirdita et al., 2021), which show a similar structure but with the additional N-terminal α-helical membrane anchor (Supplementary Figure S4).”