One contact for every twelve residues allows robust and accurate topology-level protein structure modeling

By David E. Kim, Frank DiMaio, R. Wang, Yifan Song, David Baker. 2013

90 citations recorded by the source corpus.

A number of methods have been described for identifying pairs of contacting residues in protein three‐dimensional structures, but it is unclear how many contacts are required for accurate structure modeling. The CASP10 assisted contact experiment provided a blind test of contact guided protein structure modeling. We describe the models generated for these contact guided prediction challenges using the Rosetta structure modeling methodology. For nearly all cases, the submitted models had the correct overall topology, and in some cases, they had near atomic‐level accuracy; for example the model of the 384 residue homo‐oligomeric tetramer (Tc680o) had only 2.9 Å root‐mean‐square deviation (RMSD) from the crystal structure. Our results suggest that experimental and bioinformatic methods for obtaining contact information may need to generate only one correct contact for every 12 residues in the protein to allow accurate topology level modeling. Proteins 2014; 82(Suppl 2):208–218. © 2013 Wiley Periodicals, Inc.

Subfields: Materials Chemistry, Molecular Biology, Computational Biology & Bioinformatics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Materials Chemistry

In Molecular Biology

In Computational Biology & Bioinformatics

Citations in the corpus

Evaluating Protein Transfer Learning with TAPE (2019, background): “, they facilitate robust modeling of full 3D protein structure [39].”
Enhancing Evolutionary Couplings with Deep Convolutional Neural Networks (2017, background): “This choice is exacerbated for long proteins because as the length, L, of a protein grows the number of contacts also grows in proportion to L; however, the number of entries in the contact map grows with L(2) (Kim et al., 2014).”
Large-scale determination of previously unsolved protein structures using evolutionary information (2015, methodology): “…distributed computing project, and the lowest energy (Rosetta all atom energy plus evolutionary restraint fit) models were recombined and further optimized using a new iterative version (see ‘Materials and methods’) of the RosettaCM hybridization protocol ( Song et al., 2013 ; Kim et al., 2014 ).”
CoinFold: a web server for protein contact prediction and contact-assisted protein folding (2016, background): “(A) The web interface for job submission has fields for job name (1), optional user email address (2), and sequences to be submitted (3).”
De Novo Protein Design for Novel Folds using Guided Conditional Wasserstein Generative Adversarial Networks (gcWGAN) (2019, background): “Speciﬁcally, sequence coevolution can be exploited to infer residue-residue structure contacts (Weigt et al. , 2009; Kamisetty et al. , 2013; Seemayer et al. , 2014) and enhance protein structure prediction signiﬁcantly (Kim et al. , 2014; Hopf et al. , 2014; Ovchinnikov et al. , 2014).”
Analysis of deep learning methods for blind protein contact prediction in CASP12 (2017, background): “Recent progress has indicated that long-range contacts are very helpful for structure prediction [1] and that direct co-evolution analysis may correctly predict some long-range contacts for proteins with a very large number of sequence homologs [2, 3].”
ConEVA: a toolbox for comprehensive assessment of protein contacts (2016, background): “For instance, DE et al. suggest that 1 contact in every 12 residues is sufficient to robustly fold a protein at topology level [35].”
DESTINI: A deep-learning approach to contact-driven protein structure prediction (2019, background): “Another study estimates that, on average, one contact per twelve residues allows for robust and accurate protein fold modelling 9 .”
Deep learning methods in protein structure prediction (2020, background): “Nonetheless, one contact every twelve residues may be sufficient to allow robust and accurate topology-level protein structure modeling [97].”