Subcellular localization for Gram positive and Gram negative bacterial proteins using linear interpolation smoothing model

• Published in: Journal of theoretical biology Vol. 386; pp. 25 - 33
• Main Authors: Saini, Harsh, Raicar, Gaurav, Dehzangi, Abdollah, Lal, Sunil, Sharma, Alok
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 07.12.2015
• Subjects: Algorithms; Bacteria; Bacterial Proteins - metabolism; Dependency models; Feature extraction; Gram-Negative Bacteria - metabolism; Gram-Positive Bacteria - metabolism; Hidden Markov models; Markov Chains; Models, Statistical; Natural Language Processing; Proteomics - methods; Sensitivity and Specificity; Subcellular Fractions - metabolism; Feature extraction; Dependency models; Natural language processing; Hidden Markov models
• ISSN: 0022-5193; 1095-8541
• DOI: 10.1016/j.jtbi.2015.08.020

Protein subcellular localization is an important topic in proteomics since it is related to a protein׳s overall function, helps in the understanding of metabolic pathways, and in drug design and discovery. In this paper, a basic approximation technique from natural language processing called the linear interpolation smoothing model is applied for predicting protein subcellular localizations. The proposed approach extracts features from syntactical information in protein sequences to build probabilistic profiles using dependency models, which are used in linear interpolation to determine how likely is a sequence to belong to a particular subcellular location. This technique builds a statistical model based on maximum likelihood. It is able to deal effectively with high dimensionality that hinders other traditional classifiers such as Support Vector Machines or k-Nearest Neighbours without sacrificing performance. This approach has been evaluated by predicting subcellular localizations of Gram positive and Gram negative bacterial proteins. •We introduce a novel classifier, linear interpolation, for subcellular localization.•Inspiration to use this technique came from natural language processing.•The techniques tries to model dependencies between amino acids.•We achieved good results on two bacterial datasets.