Prediction of protein folding rates from primary sequences using hybrid sequence representation

Abstract

The ability to predict protein folding rates constitutes an important step in understanding the overall folding mechanisms. Although many of the prediction methods are structure based, successful predictions can also be obtained from the sequence. We developed a novel method called prediction of protein folding rates (PPFR), for the prediction of protein folding rates from protein sequences. PPFR implements a linear regression model for each of the mainstream folding dynamics including two-, multi-, and mixed-state proteins. The proposed method provides predictions characterized by strong correlations with the experimental folding rates, which equal 0.87 for the two- and multistate proteins and 0.82 for the mixed-state proteins, when evaluated with out-of-sample jackknife test. Based on in-sample and out-of-sample tests, the PPFR's predictions are shown to be better than most of other sequence only and structure-based predictors and complementary to the predictions of the most recent sequence-based QRSM method. We show that simultaneous incorporation of several characteristics, including the sequence, physiochemical properties of residues, and predicted secondary structure provides improved quality. This hybridized prediction model was analyzed to reveal the complementary factors that can be used in tandem to predict folding rates. We show that bigger proteins require more time for folding, higher helical and coil content and the presence of Phe, Asn, and Gln may accelerate the folding process, the inclusion of Ile, Val, Thr, and Ser may slow down the folding process, and for the two-state proteins increased beta-strand content may decelerate the folding process. Finally, PPFR provides strong correlation when predicting sequences with low similarity.

Predicting protein folding rates using the concept of Chou's pseudo amino acid composition

“…The correlation in Figure 2 may be improved if these proteins are removed from the dataset as carried in previous studies. 22,23,[25][26][27] (2) Only one kind of physicochemical property, hydrophobicity, was used in this approach. More sequence order information would be involved if we selected more than one type of physicochemical property.…”

“…This feature selection method has also been successfully applied to design the PPFR method. 25 The aim of this procedure is to ensure that the selected features do not overfit the dataset and improve the predicted accuracy.…”

“…From the literatures, [13][14][15][20][21][22][23][24][25][32][33][34][35] we have collected 117 proteins with known experimentally determined folding rates. Protein sequence homology in datasets is known to influence the prediction accuracy, i.e., the prediction accuracy will be overestimated when using highly homologous protein sequences.…”

Proteome-Level Interplay between Folding and Aggregation Propensities of Proteins

“…1,2 Indeed, substantial progress has been made in the prediction of native-state structures of proteins by using only the information provided by their sequences, so that it is currently possible to generate models for the structures of small globular proteins with a relatively high degree of confidence and accuracy, as shown by the increasing quality of the results of the CASP exercise. 3 It has also been shown that from the knowledge of the amino acid sequence of a protein it is possible to predict its folding rate; [4][5][6][7][8] in principle, the folding pathway of a protein should be predictable by just considering its sequence, at least after deriving from the sequence itself a model of the native state topology. 9,10 It has also been established that the amino acid sequences of proteins determine, to a large extent, also their aggregation behavior.…”

Real value prediction of protein folding rate change upon point mutation

“…Further, several methods have been reported to predict protein folding rates from amino acid sequence. These methods include the relationship with amino acid properties [17][18][19][20], predicted secondary structures [21], predicted inter-residue contacts [22], amino acid composition [23,24], secondary structure length [25], hybrid sequence representation [26], and flexibility and solvent accessibility [27]. The details of computational methods for predicting protein folding rates have been reviewed recently [11,28].…”