Denaturing gradient gel electrophoresis (DGGE) is one of the most commonly used molecular tools to study complex microbial communities. Despite its widespread use, meaningful interpretative analysis remains a major drawback of this method. We evaluated the combination of computer-assisted band-matching with nonparametric statistics for comparative analysis of DGGE banding patterns. Fecal samples from 17 healthy volunteers who consumed 20 g of the prebiotic compound oligofructose-enriched inulin (OF-IN) for 4 weeks were analyzed before and after treatment. DGGE fingerprinting profiles were analyzed using bionumerics software version 4.6., which resulted in a data matrix that was used for statistical analysis. When comparing DGGE profiles before and after OF-IN intake with a Wilcoxon nonparametric test for paired data, two band-classes increased significantly after OF-IN intake (P<0.003 and <0.02). These two band-classes could be assigned to the species Bifidobacterium longum and Bifidobacterium adolescentis by band-sequencing analysis, and their significant increase was quantitatively confirmed with real-time PCR using species-specific primers (respectively P<0.012 and <0.010). Therefore, the nonparametric analysis of a data matrix obtained by computer-assisted band-matching of complex profiles facilitated the interpretative analysis of these profiles and provided an objective and high-throughput method for the detection of significant taxonomic differences in larger numbers of complex profiles. © 2010 Federation of European Microbiological Societies. Published by Blackwell Publishing Ltd. All rights reserved.
