2011 | OriginalPaper | Chapter
Taxonomical Classification of Closely Related Reads of Genus Bacillus
Author : Wenmin Wang
Published in: Informatics Engineering and Information Science
Publisher: Springer Berlin Heidelberg
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
The genus Bacillus contain spore-forming gram-positive/variable rod-shaped bacteria. Species of the Bacillus genus have long believed to have medical, veterinary and agricultural importance. In agricultural biotechnology and its applications, discriminating short environmental Bacillus DNA fragments into its various species members plays a crucial role in the pipeline of agronomic trait discovery and insect control. We here constructed a classification model for this challenging task based on consensus decision-making of support vector machines and BLAST hit strategies. We first took advantage of both the hexamer signatures of Bacillus genomes and the Bacillus species-specific toxin signatures to build the attribute space. We then explored and filtered the otherwise high dimensional attribute space with a weighted version of principal component analysis to mitigate computational cost and avoid possible overfitting of the classification model for discriminating Bacillus species. Our extensive experimental results showed that our method can perform well on differentiating Bacillus species.