Compared with culture-independent approaches, traditionally used culture-dependent methods have a limited capacity to characterize water microbiota. Nevertheless, for almost a century the latter have been optimized to detect and quantify relevant bacteria. A pertinent question is if culture-independent diversity surveys give merely an extended perspective of the bacterial diversity or if, even with a higher coverage, focus on a different set of organisms. We compared the diversity and phylogeny of bacteria in a freshwater sample recovered by currently used culture-dependent and culture-independent methods (DGGE and 454 pyrosequencing). The culture-dependent diversity survey presented lower coverage than the other methods. However, it allowed bacterial identifications to the species level, in contrast with the other procedures that rarely produced identifications below the order. Although the predominant bacterial phyla detected by both approaches were the same (Proteobacteria, Actinobacteria, Bacteroidetes), sequence similarity analysis showed that, in general, different operational taxonomical units were targeted by each method. The observation that culture-dependent and independent approaches target different organisms has implications for the use of the latter for studies in which taxonomic identification has a predictive value. In comparison to DGGE, 454 pyrosequencing method had a higher capacity to explore the bacterial richness and to detect cultured organisms, being also less laborious.