A conditional neural fields model for protein threading

By Jianzhu Ma, Jian Peng, Sheng Wang, Jinbo Xu. 2012

87 citations recorded by the source corpus.

Motivation: Alignment errors are still the main bottleneck for current template-based protein modeling (TM) methods, including protein threading and homology modeling, especially when the sequence identity between two proteins under consideration is low (<30%). Results: We present a novel protein threading method, CNFpred, which achieves much more accurate sequence–template alignment by employing a probabilistic graphical model called a Conditional Neural Field (CNF), which aligns one protein sequence to its remote template using a non-linear scoring function. This scoring function accounts for correlation among a variety of protein sequence and structure features, makes use of information in the neighborhood of two residues to be aligned, and is thus much more sensitive than the widely used linear or profile-based scoring function. To train this CNF threading model, we employ a novel quality-sensitive method, instead of the standard maximum-likelihood method, to maximize directly the expected quality of the training set. Experimental results show that CNFpred generates significantly better alignments than the best profile-based and threading methods on several public (but small) benchmarks as well as our own large dataset. CNFpred outperforms others regardless of the lengths or classes of proteins, and works particularly well for proteins with sparse sequence profiles due to the effective utilization of structure information. Our methodology can also be adapted to protein sequence alignment. Contact: j3xu@ttic.edu Supplementary information: Supplementary data are available at Bioinformatics online.

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

Distance-based protein folding powered by deep learning (2018, methodology): “We further compare RaptorX-Contact with 2 in-house threading methods, CNFpred (36, 37) and RaptorX-TBM.”
MRFalign: Protein Homology Detection through Alignment of Markov Random Fields (2014, methodology): “We use a Conditional Neural Fields (CNF) [40] method, which is very similar to what is described in the protein threading paper [41], to estimate the occurring probability of an alignment and then derive node alignment potential from this CNF.”
RaptorX-Property: a web server for protein structure property prediction (2016, methodology): “When only one hidden layer is used, DeepCNF becomes conditional neural fields (CNF) (14).”
Protein threading using context-specific alignment potential (2013, unknown): “We use a recently developed probabilistic graphical model conditional neural fields (Peng et al., 2009) to calculate P(AjS,T) as follows (Ma et al., 2012), P AjT, S, ð Þ ¼ exp F AjT, S, ð Þ þ wG AjT, S, ð Þð Þ ZðT, S, Þ ð2Þ where is the model parameter vector to be trained, w (¼ 1.0) is a weight…”
Fully Connected Deep Structured Networks (2015, background): “More recently [5, 27, 22, 6, 28, 25] incorporate non-linearities into unary potentials but generally assume exact inference to be tractable.”
DeepCNF-D: Predicting Protein Order/Disorder Regions by Weighted Deep Convolutional Neural Fields (2015, background): “The order/disorder state of a residue has a strong relationship with the residue’s substitution and evolution [46–48].”
Protein Homology Detection Through Alignment of Markov Random Fields (2015, methodology): “We use a Conditional Neural Fields (CNF) [12] method to develop such a node alignment potential, using a procedure very similar to what is described in the protein threading paper [13, 14].”