Ortho_Sim_Loc: Essential protein prediction using orthology and priority-based similarity approach. (June 2021)

• Record Type: Journal Article
• Authors: Payra, Anjan Kumar; Saha, Banani; Ghosh, Anupam
• Is Part Of: Computational biology and chemistry. Volume 92(2021)
• Publication Date: 2021-06
• Subjects: Classifier -- Clustering -- Sub-cellular localization -- Orthologous groups -- Protein-protein interaction -- 3-sigma; Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2021.107503
• Languages: English
• ISSNs: 1476-9271

Abstract

Graphical abstract: Flowchart of essential protein prediction using priority based similarity measures. Highlights: Matrix of essential properties is constructed using unbiased approach of classifiers. Cluster is selected using cumulative GO-Annotation score to find functionally enriched group of proteins. Association scores is used between different orthologous groups (COG). The results are validated using GO-Attributes (Functionally enriched set of proteins) counts. The proposed methodology has achieved higher success rates in comparison to the existing methods. Abstract: Proteins are the essential macro-molecules of living organism. But all proteins cannot be considered as essential in different relevant studies. Essentiality of a protein is thus computed by computation methods rather than biological experiments which in turn save both time and effort. Different computational approaches are already predicted to select essential proteins successfully with different biological significances by researchers. Most of the experimental approaches return higher false negative outcomes with respect to others. In order to retain the prediction accuracy level, a novel methodology " Ortho_Sim_Loc" has been proposed which is a combined approach of Orthology, Similarity (using clustering and priority based GO-Annotation) and Subcellular localization. Ortho_Sim_Loc can predict enriched functional set essential proteins. The predicted results are validated with other existing methods like different centrality measures, LIDC. The validation results exhibits better performance of O r t h o _ S i m _ L o c in compare to other existing computational approaches.