Annotation of Bacteria by Greengenes Classifier Using 16S rRNA Gene Hyper Variable Regions

Abstract

rRNA-genes for phylogenetic classifications started to be used in 1980s first time by Carl Woese which made a ground breaking contribution to microbiome science. rRNA-genes are used to explore microbial diversity as well as a method for bacterial annotation. Many researchers followed rRNA-based analysis track as a central method in microbiology. Similarity based analyses use several new generations of Artificial Neural Networks to create classifiers against bacteria libraries to obtain high accuracies. By the time, the number of bacteria in these libraries increased enormously. In this article the accuracy of a classifier against Greenges library is tested. It has been shown by the author in previous articles that the Greengenes Classifier can be successfully used as a bioinformatics program that performs taxonomic classification of 16S rRNA gene sequences. In a previous article, the accuracy of the program is also tested when it is applied to common PCR products of the 16S rRNA variable regions, which are the only product of laboratories in microbiome projects. In this study, V1–V3 hyper variable regions from 16S rRNA genes of some known bacteria is taken from the work of A. Cosic. In this article we used Longest Common Subsequence similarity measure to classify bacterial 16S rRNA gene sequence short reads against the Greengenes library.