iPhos‐PseEvo: Identifying Human Phosphorylated Proteins by Incorporating Evolutionary Information into General PseAAC via Grey System Theory

Abstract

Protein phosphorylation plays a critical role in human body by altering the structural conformation of a protein, causing it to become activated/deactivated, or functional modification. Given an uncharacterized protein sequence, can we predict whether it may be phosphorylated or may not? This is no doubt a very meaningful problem for both basic research and drug development. Unfortunately, to our best knowledge, so far no high throughput bioinformatics tool whatsoever has been developed to address such a very basic but important problem due to its extremely complexity and lacking sufficient training data. Here we proposed a predictor called iPhos-PseEvo by (1) incorporating the protein sequence evolutionary information into the general pseudo amino acid composition (PseAAC) via the grey system theory, (2) balancing out the skewed training datasets by the asymmetric bootstrap approach, and (3) constructing an ensemble predictor by fusing an array of individual random forest classifiers thru a voting system. Rigorous jackknife tests have indicated that very promising success rates have been achieved by iPhos-PseEvo even for such a difficult problem. A user-friendly web-server for iPhos-PseEvo has been established at http://www.jci-bioinfo.cn/iPhos-PseEvo, by which users can easily obtain their desired results without the need to go through the complicated mathematical equations involved. It has not escaped our notice that the formulation and approach presented here can be used to analyze many other problems in protein science as well.

iHyd-PseCp: Identify hydroxyproline and hydroxylysine in proteins by incorporating sequence-coupled effects into general PseAAC

“…Therefore, using Eq 9. has made the meanings of sensitivity, specificity, overall accuracy, and Mathew's correlation coefficient much more intuitive and easier-to-understand, particularly for the meaning of MCC, as concurred recently by many investigators (see, e.g., [11, 12, 16, 18, 19, 21–23, 64, 65, 99–108]).…”

“…It is instructive to point out that, of the four metrics, the most important are the Acc and MCC [11, 12, 21, 22]: the former reflects the overall accuracy of a predictor; while the latter, its stability in practical applications. The metrics Sn and Sp are used to measure a predictor from two opposite angles.…”

Assessing the Performances of Protein Function Prediction Algorithms from the Perspectives of Identification Accuracy and False Discovery Rate

“…To develop a useful sequence-based statistical predictor for a biological system as reported in a series of recent publications [74][75][76][77][78][79][80][81][82][83], the Chou's 5-step rule should be observed [84]: (1) How to construct or select a valid dataset to train and test the predictor? (2) How to formulate the biological sequence samples with an effective mathematical expression that can truly reflect their intrinsic correlation with the target to be predicted?…”

Pse-in-One 2.0: An Improved Package of Web Servers for Generating Various Modes of Pseudo Components of DNA, RNA, and Protein Sequences

“…This is because almost all the existing machine-learning algorithms, such as "Neural Network" or NN algorithm [1][2][3] "Support Vector Machine" or SVM algorithm [4][5][6][7][8][9][10][11][12] "Nearest Neighbor" or NN algorithm [13,14] and "Random Forest" algorithm [15][16][17][18][19][20][21][22] can only handle vectors but not sequence samples as elucidated in a review paper [23]. Unfortunately, if using the sequential model, i.e., the model in which all the samples are represented by their original sequences, it is hardly able to train a machine learning model that can cover all the possible cases concerned, as elaborated in [24].…”