One-dimensional metric for tracking bacterial community divergence using sole carbon source utilization patterns

Abstract

Community level physiological profiling (CLPP) has become a popular method to characterize and track changes in heterotrophic bacterial communities. Although the CLPP method is a straight forward laboratory protocol which yields large amounts of functional information, the amount of data obtained can become overwhelming and often requires some type of multivariate analysis method for ordination and interpretation. Multivariate analysis can be challenging and requires a significant statistics background along with an understanding of the inferences and biases each multivariate analysis method incurs. This paper presents and evaluates a new approach to analyzing sole carbon source utilization data. A method is described which provides a one-dimensional metric derived from standard CLPP data (BIOLOG™ EcoPlate data). The one-dimensional community metric was derived using normalized Euclidean distances and shifts in the carbon source utilization patterns. The one-dimensional community metric did not provide all of the information of classical approaches such as principle component analysis (PCA) or guild grouping analysis; however, it was found to be more easily implemented and interpreted when analyzing the plate data. Validation of this approach is demonstrated using data acquired to track the divergence of bacterial communities in wetland mesocosm systems after an experimentally controlled disturbance. If the objective is to investigate community shifts over time the one-dimensional community divergence metric can be a useful tool.