An Effective Multi-Label Protein Sub-Chloroplast Localization Prediction by Skipped-grams of Evolutionary Profiles using Deep Neural Network

Abstract

Chloroplast is one of the most classic organelles in algae and plant cells. Identifying the locations of chloroplast proteins in the chloroplast organelle is an important as well as a challenging task in deciphering their functions. Biological experiments to identify the protein sub-chloroplast localization (PSCL) is time-consuming and cost-intensive. Over the last decade, a few computational methods have been developed to predict PSCL in which earlier works assumed to predict only single-location; whereas, recent works are able to predict multiple-locations of chloroplast organelle. However, the performances of all the state-of-the-art predictors are poor. This study proposes a novel skipped gram technique to extract high discriminating patterns from evolutionary profiles and a multi-label deep neural network is proposed to predict the PSCL. The proposed model is assessed on two publicly available stringent datasets, i.e., Benchmark and Novel. Experimental results demonstrate that the proposed model's performance significantly outperforms in all the evaluation metrics when compared to the multi-label state-of-the-art predictors. The proposed model's multi-label accuracy (i.e., Overall Actual Accuracy) is enhanced with respect to the best PSCL predictor from the literature by a minimum margin of 6.7% (absolute) on Benchmark and 7.9% (absolute) on Novel datasets. IEEE