Identification of discriminative characteristics for clusters from biologic data with InforBIO software

Background: There are a number of different methods for generation of trees and algorithms for phylogenetic analysis in the study of bacterial taxonomy. Genotypic information, such as SSU rRNA gene sequences, now plays a more prominent role in microbial systematics than does phenotypic information. However, the integration of genotypic and phenotypic information for polyphasic studies is necessary for the classification and identification of microbes. Thus, we devised an algorithm that objectively identifies discriminative characteristics for focused clusters on generated trees from a dataset composed of coded data, such as phenotypic information. Moreover, this algorithm has been integrated into the polyphasic analysis software, InforBIO. Results: We developed a differential-character-finding algorithm based on information measures and used this algorithm to identify the characteristic that best discriminates operational taxonomic unit clusters. For all characteristics in a dataset, ...

• Type: Journal
• Year: 2007
• Where: BMCBI
• Authors: Naoto Tanaka, Masataka Uchino, Satoru Miyazaki, Hideaki Sugawara
• Keywords: Algorithm, BMCBI 2007, Discriminative Characteristics, Phenotypic Information