Improved protein contact predictions with the MetaPSICOV2 server in CASP12

By Daniel W. A. Buchan, David T. Jones. 2017

54 citations recorded by the source corpus.

In this paper, we present the results for the MetaPSICOV2 contact prediction server in the CASP12 community experiment (http://predictioncenter.org). Over the 35 assessed Free Modelling target domains the MetaPSICOV2 server achieved a mean precision of 43.27%, a substantial increase relative to the server's performance in the CASP11 experiment. In the following paper, we discuss improvements to the MetaPSICOV2 server, covering both changes to the neural network and attempts to integrate contact predictions on a domain basis into the prediction pipeline. We also discuss some limitations in the CASP12 assessment which may have overestimated the performance of our method.

Subfields: Molecular Biology

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

Citations in the corpus

Analysis of several key factors influencing deep learning-based inter-residue contact prediction (2019, background): “3 Impact of domain-based contact prediction It has been shown that integrating domain-level contact predictions with full-length contacts can improve contact prediction precision, especially when domain boundaries can be accurately identified (Buchan and Jones, 2018).”
Modeling aspects of the language of life through transfer-learning protein sequences (2019, background): “Despite recent successes [49, 50], this is still not possible for all proteins.”
Assessment of Contact Predictions in CASP12: Co-Evolution and Deep Learning Coming of Age (2017, background): “Some predictors indicate that improved methodology including deeper networks is one of the main drivers for the significant improvement in predictions.(48,49) Another reason for the enhanced performance in CASP12 appears to be the increase in the number of sequences available (Figure 16) and, thus, ”
Assessment of Prediction Methods for Protein Structures Determined by NMR in CASP14: Impact of AlphaFold2 (2021, unknown): “1 These breakthrough technologies exploit advances in attention-based machine learning, 2,3 contact prediction based on sequence co-variance analysis using the massive data bases of genomic sequence data, 4 – 9 and the rapidly growing database of experimental protein structures.”
Deep learning methods in protein structure prediction (2020, background): “DNCON2 is a two-stage CNN trained on a set of input features similar to MetaPSICOV [117].”
AlphaFold Models of Small Proteins Rival the Accuracy of Solution NMR Structures (2022, background): “, 2019), contact prediction based on sequence covariance (Marks et al., 2011; Morcos et al., 2011; Marks et al., 2012; Ovchinnikov et al., 2015; Ovchinnikov et al., 2016; Buchan and Jones, 2018), the massive and still growing databases of genomic sequence data, and the rapidly growing Protein Data B”
Assessing the accuracy of contact predictions in CASP13 (2019, methodology): “After 2010, when first papers on correlated mutations were published, a number of refinements took place, such as combining a set of nonoverlapping contact predictions in a consensus approach and especially the application of various supervised deep neural network based approaches,(42) further impro”
Ensembling multiple raw coevolutionary features with deep residual neural networks for contact-map prediction in CASP13 (2019, methodology): “Supervised machine learning based methods, such as MetaPSICOV 14,15 and NeBcon, 16 predict contact-maps by combining features based on the final results of various coevolutionary methods with a variety of one-dimensional sequence properties.”
Folding non-homologous proteins by coupling deep-learning contact maps with I-TASSER assembly simulations (2021, methodology): “, 2014), and onemeta-server-basedmethods MetaPSICOV2 (Buchan and Jones, 2018).”
Recent developments in deep learning applied to protein structure prediction (2019, methodology): “39 These results suggested that most of the information needed to accurately predict contacts between a pair of residues lies in the residue covariation data for seven or so residues on either side of the residue pair being considered.”