New encouraging developments in contact prediction: assessment of the CASP11 results

By Bohdan Monastyrskyy, Daniel D'Andrea, Krzysztof Fidelis, Anna Tramontano, Andriy Kryshtafovych. 2015

95 citations recorded by the source corpus.

This article provides a report on the state‐of‐the‐art in the prediction of intra‐molecular residue‐residue contacts in proteins based on the assessment of the predictions submitted to the CASP11 experiment. The assessment emphasis is placed on the accuracy in predicting long‐range contacts. Twenty‐nine groups participated in contact prediction in CASP11. At least eight of them used the recently developed evolutionary coupling techniques, with the top group (CONSIP2) reaching precision of 27% on target proteins that could not be modeled by homology. This result indicates a breakthrough in the development of methods based on the correlated mutation approach. Successful prediction of contacts was shown to be practically helpful in modeling three‐dimensional structures; in particular target T0806 was modeled exceedingly well with accuracy not yet seen for ab initio targets of this size (>250 residues). Proteins 2016; 84(Suppl 1):131–144. © 2015 Wiley Periodicals, Inc.

Subfields: Mechanical Engineering, Mechanics of Materials, 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 Mechanical Engineering

In Mechanics of Materials

In Computational Biology & Bioinformatics

Citations in the corpus

DEEPCON: Protein Contact Prediction using Dilated Convolutional Neural Networks with Dropout (2019, background): “As defined in the Critical Assessment of Protein Structure Prediction (CASP) experiments [4,5], a pair of”
Critical Assessment of Methods of Protein Structure Prediction (CASP) – Progress and New directions in Round XI (2016, background): “13 The possibility of predicting three dimensional long range residue–residue contacts from evolutionary information was first recognized about the time CASP started in 1994,14 and because of its promise, an assessment area on this topic was introduced early in the history of CASP.”
Folding membrane proteins by deep transfer learning (2017, methodology): “CMAPpro (Di Lena et al., 2012) uses a deep architecture, but its performance was worse than MetaPSICOV in CASP11 (Monastyrskyy et al., 2015).”
Deep learning extends de novo protein modelling coverage of genomes using iteratively predicted structural constraints (2018, background): “In recent years, the ability to accurately predict residue-residue contacts in protein structures from a family of protein sequences has increased dramatically, mainly due to the recent breakthrough in developing statistical models which can separate direct from indirect correlation effects [1,2].”
Assessing the accuracy of contact predictions in CASP13 (2019, background): “After a long hiatus, in 2014 at the CASP11 meeting the first reports and signs of advance appeared in the accuracy of contact prediction, 5 these were followed by a nearly doubling in the precision of prediction accuracy at CASP12 4 in 2016.”
ConEVA: a toolbox for comprehensive assessment of protein contacts (2016, methodology): “Besides these “ reduced list ” metrics [32] that only evaluate selected top contacts, ConEVA also presents “ full list ” metrics including Matthew ’ s correlation coefficient (MCC), area under the precision-recall curve (AUC_PR) [30], and Receiver Operating Characteristic (ROC) curve.”
CONFOLD2: improved contact-driven ab initio protein structure modeling (2017, background): “Methods that can build models faster and are more residue contact sensitive are needed to realize the promise of ab initio protein structure prediction driven by the recent advances in contact prediction [1, 2].”