Fast pseudolikelihood maximization for direct-coupling analysis of protein structure from many homologous amino-acid sequences

By Magnus Ekeberg, Tuomo Hartonen, Erik Aurell. 2014

177 citations recorded by the source corpus.

Abstract Direct-coupling analysis is a group of methods to harvest information about coevolving residues in a protein family by learning a generative model in an exponential family from data. In protein families of realistic size, this learning can only be done approximately, and there is a trade-off between inference precision and computational speed. We here show that an earlier introduced l 2 -regularized pseudolikelihood maximization method called plmDCA can be modified as to be easily parallelizable, as well as inherently faster on a single processor, at negligible difference in accuracy. We test the new incarnation of the method on 143 protein family/structure-pairs from the Protein Families database (PFAM), one of the larger tests of this class of algorithms to date.

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

Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model (2016, background): “However, for many proteins especially those without many sequence homologs, the predicted contacts by the state-of-the-art predictors such as CCMpred [4], PSICOV [5], Evfold [6], plmDCA[7], Gremlin[8], MetaPSICOV [9] and CoinDCA [10] are still of low quality and insufficient for accurate contact-ass”
Accurate contact predictions using covariation techniques and machine learning (2015, background): “Therefore, like other authors,7,18,19 we calculate effective sequence counts ( N eff ) rather than raw sequence numbers to account for this redundancy.”
Coevolutionary Analysis of Protein Subfamilies by Sequence Reweighting (2019, methodology): “Direct-Coupling Analysis (DCA) was performed using the asymmetric pseudo-likelihood inference method [37,40] as implemented in the lbsDCA package [39] using default inference parameters (available at https://gitlab.”
Inferring Contacting Residues within and between Proteins: What Do the Probabilities Mean? (2016, methodology): “It appears to me that it would be straightforward to adapt the pseudo-likelihood approach presented in [34] to calculate, for any possible graph G, the optimal pseudo-likelihood that can be obtained using the model (3), with the constraints E ab 1⁄4 0 whenever (i, j) are not contacting in G.”
Inferring interaction partners from protein sequences (2016, background): “[39] Ekeberg M, Hartonen T, Aurell E (2014) Fast pseudo-”
I-TASSER-MTD: a deep-learning-based platform for multi-domain protein structure and function prediction (2022, methodology): “…sequence, three 1D features (i.e., ﬁ eld parameters of the Potts model, a one-hot representation of the sequence, and a hidden Markov model) and two 2D coevolutionary features (i.e., the pseudolikelihood maximized Potts model 56 and mutual information) are extracted from the MSA created by DeepMSA.”
Learning protein constitutive motifs from sequence data (2018, methodology): “Representative weights of the protein families selected in Ekeberg et al. (2014) ??. RBM parameters: 21 = 0.25,M = 0.05 ×N × 20.”
Ensembling multiple raw coevolutionary features with deep residual neural networks for contact-map prediction in CASP13 (2019, methodology): “…and ResTriplet, are proposed to ensemble a triplet of raw coevolutionary features, including the covariance matrix, the precision matrix, and the parameter matrix of a pseudo-likelihood maximized Potts model, 25,26 by two complementary strategies, based on end-to-end training and stacking.”
rawMSA: End-to-end Deep Learning Makes Protein Sequence Profiles and Feature Extraction obsolete (2018, background): “More recently, contact map prediction methods have been at the center of renewed interest after the development of a number of techniques to analyse MSAs in search of direct evolutionary couplings, which have brought along a big leap in the state of the art [38–40].”