Statistical comparisons of community catabolic profiles

Comparative catabolic profiling using microtiterplates with multiple sole-carbon-sources has become a popular tool for the comparison of microbial communities with respect to their functional potential. However, the statistical treatment of such data has only been partially satisfactory so far. In this paper, a new multivariate approach developed by Lauter and coworkers [J. Lauter, Exact T and F tests for analysing studies with multiple endpoints, Biometrics 52 (1996) 964-970; J. Lauter, E. Glimm, S. Kropf, New multivariate tests for data with an inherent structure, Biometrical Journal 38 (1996) 5-23.] is applied to this problem. Using replicate observations from every community in question, this approach allows testing for significant differences between communities. It is advantageous with respect to power and maintains the alpha-level. Discriminating carbon-sources are found according to weighted factor loadings and univariate tests. Exemplarily, changes in microbial soil communities due to rotting of plants are analyzed. Furthermore, other statistical approaches from the literature are discussed. (C) 1997 Elsevier Science B.V.