More, Ravi and Purohit, H J (2016) The Identification of Discriminating Patterns from 16S rRNA Gene to Generate Signature for Bacillus Genus. Journal of Computational Biology, 23 (8). pp. 1-11. ISSN 1066-5277, ESSN: 1557-8666

[img] PDF - Published Version
Restricted to Registered users only

Download (646Kb) | Request a copy


The 16S ribosomal RNA (16S rRNA) gene has been widely used for the taxonomic classi- fication of bacteria. A molecular signature is a set of nucleotide patterns, which constitute a regular expression that is specific to each particular taxon. Our main goal was to identify discriminating nucleotide patterns in 16S rRNA gene and then to generate signatures for taxonomic classification. To demonstrate our approach, we used the phylum Firmicutes as a model using representative taxa Bacilli (class), Bacillales (order), Bacillaceae (family), and Bacillus (genus), according to their dominance at each hierarchical taxonomic level. We applied combined composite vector and multiple sequence alignment approaches to generate gene-specific signatures. Further, we mapped all the patterns into the different hypervariable regions of 16S rRNA gene and confirmed the most appropriate distinguishing region as V3–V4 for targeted taxa. We also examined the evolution in discriminating patterns of signatures across taxonomic levels. We assessed the comparative classification accuracy of signatures with other methods (i.e., RDP Classifier, KNN, and SINA). Results revealed that the signatures for taxa Bacilli, Bacillales, Bacillaceae, and Bacillus could correctly classify isolate sequences with sensitivity of 0.99, 0.97, 0.94, and 0.89, respectively, and specificity close to 0.99. We developed signature-based software DNA Barcode Identification (DNA BarID) for taxonomic classification that is available at website DNA_BarID.htm. This pattern-based study provides a deeper understanding of taxonspecific discriminating patterns in 16S rRNA gene with respect to taxonomic classification.

Item Type: Article
Uncontrolled Keywords: Computational Molecular Biology; Evolution; Learning; Sequence Analysis
Subjects: Microbiology
Depositing User: Dr. H J Purohit
Date Deposited: 03 Apr 2017 07:26
Last Modified: 03 Apr 2017 07:26

Actions (login required)

View Item View Item