Development of Computational Resources for Predicting Disease Resistance Genes and Mapping NGS Transcripts to Secondary Metabolism in Plants

Abstract

Plant disease outbreak is increasing rapidly around the globe and is a major cause for crop loss worldwide. Plants, in turn, have developed diverse defense mechanisms to identify and evade different pathogenic microorganisms. Early identification of plant disease resistance genes (R genes) can be exploited for crop improvement programs. The existing prediction methods are either based on sequence similarity/domain-based methods or electronically annotated sequences, which might miss existing unrecognized proteins or low similarity proteins. Therefore, there was an urgent need to devise a novel machine learning tool to address this problem. Considering these gaps and importance of disease resistance genes, DRPPP (Disease resistance plant protein predictor), a support vector machine (SVM) learning based tool was developed. 16 different methods were generated through feature extraction method and were employed to generate 10,270 features. Radial basis function was used and ten-fold cross validation was performed to optimize SVM parameters. The model for DRPPP was derived using LibSVM and achieved an overall accuracy of 91.11% on the test dataset. The tool was found to be robust and can be used for high-throughput datasets. newlineFurthermore, in plants the importance of medicinal herbs can be derived from the fact that the demand for herbal medicines is estimated to increase upto US$3 trillion by 2020. However, their genetic improvement has been hampered due to lack of genome resources. Next-generation sequencing (NGS) has provided unprecedented opportunities for high throughput research on medicinal plants, especially for those whose genome/transcriptome datasets were still not available. Therefore, NGS transcriptomes for two critically endangered species i.e. Aconitum heterophyllum and Swertia chirayita having high therapeutic values were generated and computationally analyzed for varying conditions of secondary metabolites. In total, four transcriptomes were generated for differential tissues (root versus shoot).