An efficient Pearson correlation based improved random forest classification for protein structure prediction techniques. (1st October 2020)

• Record Type: Journal Article
• Authors: Kalaiselvi, B.; Thangamani, M.
• Is Part Of: Measurement. Volume 162(2020)
• Publication Date: 2020-10-01
• Subjects: Amino acid features -- Improved random forest classification -- Protein structure -- Weighted covariance -- Weighted mean -- Weighted pearson correlation; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2020.107885
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: Weighted pearson correlation based improved random forest classification technique. Wpc-irfc technique for protein structure prediction. Improved random forest classification. Comparative results of true positive rate using two dataset. Abstract: In biochemistry, the protein structure prediction from the primary sequence is a significant issue. Few research works are intended for performing protein structure prediction with assist of diverse data mining techniques. However, the existing technique does not provide enhanced performance for protein structure prediction. To resolve this limitation, Weighted Pearson Correlation based Improved Random Forest Classification (WPC-IRFC) Technique is introduced. The WPC-IRFC Technique is developed for enhancing the protein structure prediction performance with higher accuracy and lesser time. The WPC-IRFC uses Weighted Pearson Correlation (WPC) to select relevant amino acid features based on weighted mean and weighted covariance. After selecting the relevant amino acid features, WPC-IRFC Technique designs an Improved Random Forest Classification (IRFC) for predicting the protein structure from a big protein dataset (DS). IRFC significantly lessens the error rate of classification with aid of iteratively reweighted least squares model to accurately identify protein structures.