Assessing the utility of coevolution-based residue–residue contact predictions in a sequence- and structure-rich era

By Hetunandan Kamisetty, Sergey Ovchinnikov, David Baker. 2013

660 citations recorded by the source corpus.

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

1. Yang Zhang
2. Jinbo Xu
3. Robin Pearce
4. Sheng Wang
5. David Baker
6. David T. Jones
7. Sergey Ovchinnikov
8. Hahnbeom Park

In Molecular Biology

1. Yang Zhang
2. Jinbo Xu
3. Sheng Wang
4. David T. Jones
5. Robin Pearce
6. Sergey Ovchinnikov
7. David Baker
8. Joe G. Greener

In Computational Biology & Bioinformatics

1. Jinbo Xu
2. David T. Jones
3. Sheng Wang
4. Yang Zhang
5. Sergey Ovchinnikov
6. Robin Pearce
7. David Baker
8. Joe G. Greener

Citations in the corpus

• Improved protein structure prediction using predicted interresidue orientations (2019, background): “We found that, unlike direct-coupling based methods such as GREMLIN, the depth of the MSA did not have a good correlation with the accuracy of the derived contacts.”
• ProGen: Language Modeling for Protein Generation (2020, methodology): “Next, we use the online server of GREMLIN [21] to extract a HH∆ score.”
• Evolutionary context-integrated deep sequence modeling for protein engineering (2020, methodology): “We refer the interested [38,58,59] for the details of the optimization.”
• Protein Structure Determination using Metagenome sequence data (2017, neutral): “for​ sep_x (0,1,2) for​ sep_y (1,2,4, 8,16,32) ini_mtx = ​initialize_matrix ​(sep_x ​, ​sep_y) for​ gap_e (-0.”
• CCMpred—fast and precise prediction of protein residue–residue contacts from correlated mutations (2014, methodology): “Modern contact prediction methods differ by their strategy in the disentangling step: the most accurate class of methods (Ekeberg et al., 2013; Kamisetty et al., 2013) such as plmDCA (Ekeberg et al., 2013) and GREMLIN (Kamisetty et al., 2013) learn the direct couplings as parameters of a Markov rand”
• Robust and accurate prediction of residue–residue interactions across protein interfaces using evolutionary information (2014, methodology): “We found previously ( Kamisetty et al., 2013 ) that contacts could be predicted with high accuracy for monomeric proteins, provided there were sufficient aligned sequences, but in such cases there was almost always already a structure of a family member from which comparative models could be built,…”
• Enhancing Evolutionary Couplings with Deep Convolutional Neural Networks (2017, methodology): “Notably, evolutionary coupling analysis methods, such as direct-coupling analysis, GREMLIN, (meta-) PSICOV, and EVFold, take an MSA as input and predict residue-residue contacts by learning an inherent graphical model structure that incorporates pairwise terms to capture evolutionary constraints amo”
• Sequence co-evolution gives 3D contacts and structures of protein complexes (2014, background): “determine the 3D structure of individual proteins (Marks et al., 2011; Morcos et al., 2011; Aurell and Ekeberg, 2012; Jones et al., 2012; Kamisetty et al., 2013), including integral membrane proteins (Hopf et al.”
• Folding membrane proteins by deep transfer learning (2017, methodology): “CCMpred has similar accuracy as Gremlin and plmDCA, but on average better than PSICOV and Evfold (maximum-entropy option).”