Multivariate Linear Regression May Assist in Determining Virulence Factors for Microbes
By American Society for Microbiology
Mar 11, 2006, 20:37

A well established statistical tool known as multivariate linear regression may offer a new approach in determining contributions of multiple virulence factors to the overall virulence of pathogenic microbes say researchers from Albert Einstein College of Medicine, Bronx, New York and Westminster College, Salt Lake City, Utah. Their findings appear in the March 2006 issue of the journal Infection and Immunity.

Virulence is defined as the capacity of a microbe to cause damage to its host. Although there is much literature discussing the contribution of virulence factors to microbial virulence, there is no designated methodology for determining the impact of individual virulence factors on overall microbial virulence. Identification of such a method could greatly contribute to vaccine development and antimicrobial strategies.

Multivariate linear regression is a statistical tool used to analyze the relative contributions of different parameters. Out of the three types of multivariate linear regression researchers identified hierarchical regression as the type most applicable to this study. It is described as entering variables in different blocks in a specific order with the order of entry resulting from theoretical or logical importance. This approach was applied to Cryptococcus neoformans and Bacillus anthracis and results showed the method to be useful in determining the relative contributions of virulence factors in pathogenesis.

"Multivariate linear regression can be used to identify the relative levels of importance of virulence factors in virulence studies, and this information can be used to prioritize antigen identification for vaccine development and the design of antimicrobial strategies that target virulence mechanisms," say the researchers.

All rights reserved by