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

Abstract

One of the most important challenges in computational and molecular biology is to understand the relationship between amino acid sequences and the folding rates of proteins. Recent works suggest that topological parameters, amino acid properties, chain length and the composition index relate well with protein folding rates, however, sequence order information has seldom been considered as a property for predicting protein folding rates. In this study, amino acid sequence order was used to derive an effective method, based on an extended version of the pseudo-amino acid composition, for predicting protein folding rates without any explicit structural information. Using the jackknife cross validation test, the method was demonstrated on the largest dataset (99 proteins) reported. The method was found to provide a good correlation between the predicted and experimental folding rates. The correlation coefficient is 0.81 (with a highly significant level) and the standard error is 2.46. The reported algorithm was found to perform better than several representative sequence-based approaches using the same dataset. The results indicate that sequence order information is an important determinant of protein folding rates.

iAMP-2L: A two-level multi-label classifier for identifying antimicrobial peptides and their functional types

“…(9)) and k = 4 for getting the optimal results. It is instructive to note that, since the concept of PseAAC was introduced in 2001 [26], it has penetrated into almost all the fields of protein attribute predictions, such as predicting metalloproteinase family [33], predicting GABA(A) receptor proteins [34], predicting enzyme subfamily classes [35], predicting allergenic proteins [36], predicting cyclin proteins [37], predicting protein structural class [38], identifying bacterial virulent proteins [39], predicting DNA-binding proteins [40], predicting protein subcellular location [41], identifying protein submitochondrial localization [42], predicting apoptosis protein subcellular localization [43], predicting outer membrane proteins [44], predicting protein quaternary structure attribute [45,46], classifying amino acids [47], predicting G-protein-coupled receptor classes [48], predicting risk type of human papillomaviruses [49], predicting cyclin proteins [37], predicting protein folding rates [50], predicting protein supersecondary structure [51], among many others. Actually, the concept of PseAAC was not only limited for protein and peptide sequences; recently it was also extended to represent the feature vectors of DNA and nucleotides [52,53], as well as other biological samples (see, e.g., [54,55]).…”

Predicting Subchloroplast Locations of Proteins Based on the General Form of Chou's Pseudo Amino Acid Composition: Approached From Optimal Tripeptide Composition

“…For a summary about its recent development and applications, see a comprehensive review [8]. Ever since the concept of PseAAC was proposed by Chou [6] in 2001, it has rapidly penetrated into almost all the fields of protein attribute prediction, such as predicting protein structural classes [37,56], predicting protein quaternary structure [76], identifying bacterial virulent proteins [52], identifying cell wall lytic enzymes [18], identifying risk type of human papillomaviruses [22], identifying DNA-binding proteins [24], predicting homo-oligomeric proteins [55], predicting protein secondary structure content [3], predicting supersecondary structure [83], predicting enzyme family and sub-family classes [54,66,82], predicting protein subcellular location [35,36,80], predicting subcellular localization of apoptosis proteins [32,35,44,19], predicting protein subnuclear location [33], predicting protein submitochondria locations [75,51], predicting G-Protein-Coupled Receptor Classes [27,53], predicting protein folding rates [28], predicting outer membrane proteins [39], predicting cyclin proteins [48], predicting GABA(A) receptor proteins [49], identifying bacterial secreted proteins [73], identifying the cofactors of oxidoreductases [77], identifying lipase types [78], identifying protease family [30], predicting Golgi protein types [17], classifying amino acids [26], among many ot...…”

Large-scale identification of potential drug targets based on the topological features of human protein–protein interaction network

“…We firstly calculate the various primary structure descriptors of proteins. There are 1767 descriptors, including amino acid composition (20 dimensions), dipeptide composition (400 dimensions), normalized Moreau-Broto autocorrelation (400 dimensions), Moran autocorrelation (400 dimensions), Geary autocorrelation (400 dimensions), composition (21 dimensions), transition (21 dimensions) and distribution (105 dimensions), which have been widely utilized to predict various attributes of proteins [26][27][28][29][30]. Secondly, we model the largest connected component as a graph containing vertices and edges.…”