Prediction of Cyclin Protein Using Two-Step Feature Selection Technique

Abstract

Cyclins are a family of proteins that regulate the cell cycle by activating cyclin-dependent kinases or a group of enzymes required in the cell cycle. Constructing a model to classify Cyclins is of importance to understand their function. It is urgent to construct a machine learning based model to identify Cyclins because of low similarity between the sequence of Cyclins. In this study, a method based on support vector machine (SVM) is developed to recognize Cyclins only using amino acid sequence information. 18 feature descriptors with a total of 13151-dimension features were extracted, and the feature dimension were reduced to 8 through feature selection technique. The reserved features show some of feature descriptors such as Autocorrelation, AAC and CTDC are important in the identification of Cyclins. Jackknife cross-validation results indicate our model would classify Cyclins with an accuracy of 91.9%, which is superior to a recent study using the same data set. Our work provides an important tool for discriminating Cyclins.

ASPIRER: a new computational approach for identifying non-classical secreted proteins based on deep learning

“…The autocorrelation feature descriptors describe the difference of amino acid sequences based on their physicochemical properties and specific structure. They have been applied in numerous previous studies, such as predicting specific protein family, oligomeric states of proteins [ 22 , 23 ] and protein–protein interactions [ 24 ].…”

Identification of cyclin protein using gradient boost decision tree algorithm

“…In recent studies, Mohabatkar et al, [14] and Sun et al, [15] used the similar dataset for training their models by using jackknife cross-validation. The accuracies of their models were 83.53% and 91.90%, respectively.…”

“…Cyclin protein have low similarity between their sequences. To date, numerous predictors have been established to classify cyclins in diverse species [14] , [15] , [107] . In this study, an advanced ensemble model was established to identify cyclins.…”

“…In prior methods, StAR [14] and other classifiers using Pseudo-amino acid composition (PseAAC) could identify cyclins with an accuracy of 83.53%. Sun et al [15] established a cyclin prediction model based on support vector machine (SVM) which could produce an accuracy of 91.90%. Although both cyclin prediction model can produce good outcomes, there is still room for further improvement by extracting more feature information.…”

A k nearest neighbours classifiers ensemble based on extended neighbourhood rule and features subsets

“…Performance of the proposed method could further be improved by using appropriate distance formula to determine the paths. Another possible way to improve performance of the method is to use the feature selection procedures as given in [59,60,61,62,63,64,65]. This could be used for selecting a set of features from the total feature space for model construction.…”