Developing SSU rDNA metagenomic profiles of aquatic microbial communities for environmental assessments

Five water samples from three sources, two municipal reservoirs in central North Carolina and Toolik Lake in Alaska, were processed to conduct a comparative survey of microbial small subunit rDNA sequences. Genomic DNA was extracted and amplified by PCR using universal SSU rDNA primers to generate 16S and 18S rDNA clone libraries and 50 clones from each library were sequenced and placed in operational taxonomic units (OTUs). Through this recovery and analysis of SSU rRNA genes, a metagenomic profile of the microbial community emerged for each environmental sample. Analyses of these profiles, including species diversity estimates and rank-abundance curves, revealed that approximately 64% of prokaryotic OTUs and 80% of eukaryotic OTUs were novel. Diversity estimates were consistent with predicted ecosystem characteristics: they were greater for the mesotrophic to eutrophic temperate lakes, than for the oligotrophic arctic lake. Sample comparisons showed that community similarity declined as geographic distance between sites increased. Real-time quantitative PCR results showed that OTUs which had been recovered from only one library were actually present in other samples, but at much lower frequencies, suggesting that many, if not most, microorganisms are cosmopolitan. Together, these results support the potential value of using the microbial community as an indicator of local environmental conditions. In other words, it may be realistic to monitor water quality using a single, comprehensive suite of microorganisms by analyzing patterns of relative abundance.     

Comparison of microbial community compositions of injection and production well samples in a long-term water-flooded petroleum reservoir

Water flooding plays an important role in recovering oil from depleted petroleum reservoirs. Exactly how the microbial communities of production wells are affected by microorganisms introduced with injected water has previously not been adequately studied. Using denaturing gradient gel electrophoresis (DGGE) approach and 16S rRNA gene clone library analysis, the comparison of microbial communities is carried out between one injection water and two production waters collected from a working block of the water-flooded Gudao petroleum reservoir located in the Yellow River Delta. DGGE fingerprints showed that the similarities of the bacterial communities between the injection water and production waters were lower than between the two production waters. It was also observed that the archaeal composition among these three samples showed no significant difference. Analysis of the 16S rRNA gene clone libraries showed that the dominant groups within the injection water were Betaproteobacteria, Gammaproteobacteria and Methanomicrobia, while the dominant groups in the production waters were Gammaproteobacteria and Methanobacteria. Only 2 out of 54 bacterial operational taxonomic units (OTUs) and 5 out of 17 archaeal OTUs in the injection water were detected in the production waters, indicating that most of the microorganisms introduced by the injection water may not survive to be detected in the production waters. Additionally, there were 55.6% and 82.6% unique OTUs in the two production waters respectively, suggesting that each production well has its specific microbial composition, despite both wells being flooded with the same injection water.     

Comparison of the structure and composition of bacterial communities from temperate and tropical freshwater ecosystems

We used a partial 16S rRNA sequencing approach to compare the structure and composition of the bacterial communities in three large, deep subalpine lakes in France with those of communities in six shallow tropical reservoirs in Burkina Faso. Despite the very different characteristics of these ecosystems, we found that their bacterial communities share the same composition in regard to the relative proportions of the different phyla, suggesting that freshwater environmental conditions lead to convergence in this composition. In the same way, we found no significant difference in the richness and diversity of the bacterial communities in France and Burkina Faso. We defined core and satellite operational taxonomic units (OTUs) (sequences sharing at least 98% identity) on the basis of their abundance and their geographical distribution. The core OTUs were found either ubiquitously or only in temperate or tropical and subtropical areas, and they contained more than 70% of all the sequences retrieved in this study. In contrast, satellite OTUs were characterized by having a more restricted geographical distribution and by lower abundance. Finally, the bacterial community composition of these freshwater ecosystems in France and Burkina Faso was markedly different, showing that the history of these ecosystems and regional environmental parameters have a greater impact on the relative abundances of the different OTUs in each bacterial community than the local environmental conditions.     

Major differences of bacterial diversity and activity inside and outside of a natural iron-fertilized phytoplankton bloom in the Southern Ocean

One of the first comparisons of a natural iron fertilized bloom with a high-nutrient low-chlorophyll (HNLC) site was undertaken during the Kerguelen ocean and plateau compared study (KEOPS) cruise. To understand better the bacteria–phytoplankton relationship in the context of natural iron fertilization, bacterial diversity and activity was investigated in the bloom and in the adjacent HNLC region by 16S rDNA clone libraries and by single strand conformation polymorphism (SSCP) analysis. Both libraries were dominated by Alphaproteobacteria , Gammaproteobacteria and the Cytophaga-Flavobacteria-Bacteroides group. Cluster analysis at 99% sequence similarity yielded several microdiverse clusters and revealed striking differences between the two libraries. In the bloom, the dominant operational taxonomic units (OTUs) were the Roseobacter NAC11-7 cluster, SAR92 and a Cytophaga-Flavobacteria-Bacteroides cluster related to the agg58 group, whereas in the HNLC region, SAR11, Roseobacter RCA and Polaribacter dominated. SSCP analysis of 16S rDNA and 16S rRNA revealed contrasting dynamics of three different Roseobacter OTUs. Roseobacter NAC11-7 and NAC11-6 had higher relative abundances and activities in the bloom compared with the HNLC site and NAC11-6 was only detected at the decline of the bloom concomitant with a shift in phytoplankton composi tion. In contrast, Roseobacter RCA was relatively abundant and active both inside and outside of the bloom. These results suggest that the different OTUs within the Roseobacter group represent functional groups that each play an important role in the cycling of carbon.     

Analysis of bacterial core communities in the central Baltic by comparative RNA–DNA-based fingerprinting provides links to structure–function relationships

Understanding structure–function links of microbial communities is a central theme of microbial ecology since its beginning. To this end, we studied the spatial variability of the bacterioplankton community structure and composition across the central Baltic Sea at four stations, which were up to 450 km apart and at a depth profile representative for the central part (Gotland Deep, 235 m). Bacterial community structure was followed by 16S ribosomal RNA (rRNA)- and 16S rRNA gene-based fingerprints using single-strand conformation polymorphism (SSCP) electrophoresis. Species composition was determined by sequence analysis of SSCP bands. High similarities of the bacterioplankton communities across several hundred kilometers were observed in the surface water using RNA- and DNA-based fingerprints. In these surface communities, the RNA- and DNA-based fingerprints resulted in very different pattern, presumably indicating large difference between the active members of the community as represented by RNA-based fingerprints and the present members represented by the DNA-based fingerprints. This large discrepancy changed gradually over depth, resulting in highly similar RNA- and DNA-based fingerprints in the anoxic part of the water column below 130 m depth. A conceivable mechanism explaining this high similarity could be the reduced oxidative stress in the anoxic zone. The stable communities on the surface and in the anoxic zone indicate the strong influence of the hydrography on the bacterioplankton community structure. Comparative analysis of RNA- and DNA-based community structure provided criteria for the identification of the core community, its key members and their links to biogeochemical functions.     

Novel and Unexpected Prokaryotic Diversity in Water and Sediments of the Alkaline,Hypersaline Lakes of the Wadi An Natrun,Egypt

The phylogenetic diversity of the bacterial and archaeal community in the water and sediments of three large lakes of the Wadi An Natrun was investigated using 16S rRNA clone libraries. The bacterial community was diverse: 769 clones formed 345 operational taxonomic units (OTUs) defined at 99% 16S rRNA sequence identity. The bacterial community in both the water and sediments of the lakes was dominated by clones affiliated with the low G + C Gram-type-positive group, alpha-proteobacteria, and Bacteroidetes, (11-39, 11-30, and 10-37% of OTUs observed, respectively), patterns that have been observed in previously described alkaline, athalassohaline systems. However, a relatively high proportion of Firmicutess-related clones in the water of the lakes and alpha-proteobacteria in the sediments was observed. The bacterial community composition of the water and sediment of the same lake and of different lakes was significantly different (p < 0.05). Operational taxonomic units related to the gamma-proteobacteria were more abundant in the sediment of Lake Fazda, whereas the sediment of Lake UmRisha was dominated by members of the delta-proteobacteria. The proportion of gamma-proteobacterial and Bacteroidetes-affiliated OTUs were predominant in the water of Lake UmRisha and differed significantly from other lake waters (chi-squared analysis, p < or = 0.01). The more oxygenated and dilute nature of Lake Hamra was reflected in its microbial community composition, with the abundance of Bacillales sequences in the water, the absence of Halanaerobiales, Clostridiales, and Archaea in the water, and the presence of representatives of more phyla such as the Actinobacteria, Spirochaetes, and Verrucomicrobia. The archaeal community composition appeared less diverse: 589 clones resulted in 198 OTUs defined at 99% 16S rRNA sequence identity, and all sequences fell into the phylum Euryarchaeota. Phylogenetic analysis showed that many of the sequences were distantly related (83-90% 16S rRNA sequence identity) to cultured and uncultured archaea, with many clones forming clusters that branched deeply within the Euryarchaeota. Forty-two and 53% of the bacterial and archaeal clones had less than 90% 16S rRNA sequence identity to previously described sequences. This indicates that the water and sediments of the Wadi An Natrun harbor a unique and novel prokaryotic diversity that is different from what has been described among other alkaline, athalassohaline lakes.     

Distinct and diverse anaerobic bacterial communities in boreal lakes dominated by candidate division OD1

Lakes have a central role in the carbon cycle of the boreal landscape. These systems typically stratify in summer and their hypolimnetic microbial communities influence burial of biogenic organic matter in sediments. The composition of bacterial communities in these suboxic habitats was studied by pyrosequencing of 16S rRNA amplicons from five lakes with variable dissolved organic carbon (DOC) concentrations. Bacterioplankton communities in the hypolimnetic waters were clearly different from the surface layer with candidate division OD1, Chlorobi and Bacteroidetes as dominant community members. Several operational taxonomic units (OTUs) affiliated with candidate division OD1 were abundant and consistently present in the suboxic hypolimnion in these boreal lakes. The overall representation of this group was positively correlated with DOC and methane concentrations. Network analysis of time-series data revealed contrasting temporal patterns but suggested similar ecological roles among the abundant OTUs affiliated with candidate division OD1. Together, stable isotope data and taxonomic classification point to methane oxidation and autotrophic denitrification as important processes in the suboxic zone of boreal lakes. Our data revealed that while hypolimnetic bacterial communities are less dynamic, they appear to be more diverse than communities from the oxic surface layer. An appreciable proportion of the hypolimnetic bacteria belong to poorly described phyla.     

Examination of two years of community dynamics in an anaerobic bioreactor using fluorescence polymerase chain reaction (PCR) single-strand conformation polymorphism analysis

The structures of the bacterial and archaeal communities in an anaerobic digester were monitored over a 2 year period. The study was performed on a fluidized bed reactor fed with vinasse. The objective was to characterize the population dynamics over a long time period under constant environmental parameters. Total bacterial and archaeal populations were measured independently by fluorescence-based polymerase chain reaction (PCR) single-strand conformation polymorphism (SSCP) analysis using an automated DNA sequencer. With the current level of accuracy, the technique was able to monitor 45 bacterial and seven archaeal 16S rDNA molecules. The community dynamics were compared with molecular inventories of the microbial community based on 16S rDNA sequences done at the beginning of the study. The six archaeal and the 22 most frequent bacterial operational taxonomic units (OTUs) identified were associated with their SSCP peak counterparts. Overall, the data indicated that, throughout the period of the study, rapid significant shifts in the species composition of the bacterial community occurred, whereas the archaeal community remained relatively stable.     

Bacterial and archaeal communities inhabiting mussels,sediment and water in Indonesian anchialine lakes

Anchialine lakes are a globally rare and unique ecosystem consisting of saline lakes surrounded by land and isolated from the surrounding marine environment. These lakes host a unique flora and fauna including numerous endemic species. Relatively few studies have, however, studied the prokaryote communities present in these lakes and compared them with the surrounding ‘open water’ marine environment. In the present study, we used a 16S rRNA gene barcoded pyrosequencing approach to examine prokaryote (Bacteria and Archaea) composition in three distinct biotopes (sediment, water and the mussel Brachidontes sp.) inhabiting four habitats, namely, three marine lakes and the surrounding marine environment of Berau, Indonesia. Biotope and habitat proved significant predictors of variation in bacterial and archaeal composition and higher taxon abundance. Most bacterial sequences belonged to OTUs assigned to the Proteobacteria. Compared to sediment and water, mussels had relatively high abundances of the classes Mollicutes and Epsilonproteobacteria. Most archaeal sequences, in turn, belonged to OTUs assigned to the Crenarchaeota with the relative abundance of crenarchaeotes highest in mussel samples. For both Bacteria and Archaea, the main variation in composition was between water samples on the one hand and sediment and mussel samples on the other. Sediment and mussels also shared much more OTUs than either shared with water. Abundant bacterial OTUs in mussels were related to organisms previously obtained from corals, oysters and the deepsea mussel Bathymodiolus manusensis. Abundant archaeal OTUs in mussels, in contrast, were closely related to organisms previously obtained from sediment.     

POLYPHASIC STUDY OF ANTARCTIC CYANOBACTERIAL STRAINS1

We isolated 59 strains of cyanobacteria from the benthic microbial mats of 23 Antarctic lakes, from five locations in two regions, in order to characterize their morphological and genotypic diversity. On the basis of their morphology, the cyanobacteria were assigned to 12 species that included four Antarctic endemic taxa. Sequences of the ribosomal RNA gene were determined for 56 strains. In general, the strains closely related at the 16S rRNA gene level belonged to the same morphospecies. Nevertheless, divergences were observed concerning the diversity in terms of species richness, novelty, and geographical distribution. For the 56 strains, 21 operational taxonomic units (OTUs, defined as groups of partial 16S rRNA gene sequences with more than 97.5% similarity) were found, including nine novel and three exclusively Antarctic OTUs. Sequences of Petalonema cf. involvens and Chondrocystis sp. were determined for the first time. The internally transcribed spacer (ITS) between the 16S and the 23S rRNA genes was sequenced for 33 strains, and similar groupings were observed with the 16S rRNA gene and the ITS, even when the strains were derived from different lakes and regions. In addition, 48 strains were screened for antimicrobial and cytotoxic activities, and 17 strains were bioactive against the gram‐positive Staphylococcus aureus, or the fungi Aspergillus fumigatus and Cryptococcus neoformans. The bioactivities were not in coincidence with the phylogenetic relationships, but rather were specific to certain strains.     

Genetic diversity of small eukaryotes in lakes differing by their trophic status

Small eukaryotes, cells with a diameter of less than 5 mum, are fundamental components of lacustrine planktonic systems. In this study, small-eukaryote diversity was determined by sequencing cloned 18S rRNA genes in three libraries from lakes of differing trophic status in the Massif Central, France: the oligotrophic Lake Godivelle, the oligomesotrophic Lake Pavin, and the eutrophic Lake Aydat. This analysis shows that the least diversified library was in the eutrophic lake (12 operational taxonomic units [OTUs]) and the most diversified was in the oligomesotrophic lake (26 OTUs). Certain groups were present in at least two ecosystems, while the others were specific to one lake on the sampling date. Cryptophyta, Chrysophyceae, and the strictly heterotrophic eukaryotes, Ciliophora and fungi, were identified in the three libraries. Among the small eukaryotes found only in two lakes, Choanoflagellida and environmental sequences (LKM11) were not detected in the eutrophic system whereas Cercozoa were confined to the oligomesotrophic and eutrophic lakes. Three OTUs, linked to the Perkinsozoa, were detected only in the Aydat library, where they represented 60% of the clones of the library. Chlorophyta and Haptophyta lineages were represented by a single clone and were present only in Godivelle and Pavin, respectively. Of the 127 clones studied, classical pigmented organisms (autotrophs and mixotrophs) represented only a low proportion regardless of the library's origin. This study shows that the small-eukaryote community composition may differ as a function of trophic status; certain lineages could be detected only in a single ecosystem.     

Identification and analysis of PCB dechlorinating anaerobic enrichments by amplification: accuracy of community structure based on restriction analysis and partial sequencing of 16S rRNA genes

The composition of polychlorinated biphenyl (PCB) dechlorinating mixed communities was analysed by restriction fragment length polymorphism of PCR amplified rDNAs (ARDRA) and partial sequencing of 16S rRNA genes amplified from PCB degrading enrichments. Restriction analysis confirms that the 16S rRNA genes amplified from PCB dechlorinating communities vary depending on the PCB congener dechlorinated. Comparison of 16S rRNA sequences to published ribosomal databases indicates that the two most abundant Operational Taxonomic Units (OTUs) appear to be species of the genus Clostridium . The amount that the amplification procedure contributed to this result was determined by varying the amplification procedure and by creating an artificial template mixture. Varying the amount of template by sixfold in the amplification did not affect the distribution of OTUs but the number of OTUs observed decreased with decreasing template concentration. Comparison of products amplified from mixtures of 16S rDNA clones indicates that the more abundant Clostridium OTU did not amplify more efficiently than those of less abundant OTUs. Hybridization to a probe designed to detect the most abundant OTUs indicates that two other OTUs are closely related to this Clostridium species.     

Species composition and cyanotoxin production in periphyton mats from three lakes of varying trophic status

In lakes, benthic micro-algae and cyanobacteria (periphyton) can contribute significantly to total primary productivity and provide important food sources for benthic invertebrates. Despite recognition of their importance, few studies have explored the diversity of the algal and cyanobacterial composition of periphyton mats in temperate lakes. In this study, we sampled periphyton from three New Zealand lakes: Tikitapu (oligotrophic), ōkāreka (mesotrophic) and Rotoiti (eutrophic). Statistical analysis of morphological data showed a clear delineation in community structure among lakes and highlighted the importance of cyanobacteria. Automated rRNA intergenic spacer analysis (ARISA) and 16S rRNA gene clone libraries were used to investigate cyanobacterial diversity. Despite the close geographic proximity of the lakes, cyanobacterial species differed markedly. The 16S rRNA gene sequence analysis identified eight cyanobacterial OTUs. A comparison with other known cyanobacterial sequences in GenBank showed relatively low similarities (91-97%). Cyanotoxin analysis identified nodularin in all mats from Lake Tikitapu. ndaF gene sequences from these samples had very low (≤ 89%) homology to sequences in other known nodularin producers. To our knowledge, this is the first detection of nodularin in a freshwater environment in the absence of Nodularia. Six cyanobacteria species were isolated from Lake Tikitapu mats. None were found to produce nodularin. Five of the species shared low (< 97%) 16S rRNA gene sequence similarities with other cultured cyanobacteria.     

Phylogenetic ecology of the freshwater Actinobacteria acI lineage

The acI lineage of freshwater Actinobacteria is a cosmopolitan and often numerically dominant member of lake bacterial communities. We conducted a survey of acI 16S rRNA genes and 16S-23S rRNA internal transcribed spacer regions from 18 Wisconsin lakes and used standard nonphylogenetic and phylogenetic statistical approaches to investigate the factors that determine acI community composition at the local scale (within lakes) and at the regional scale (across lakes). Phylogenetic reconstruction of 434 acI 16S rRNA genes revealed a well-defined and highly resolved phylogeny. Eleven previously unrecognized monophyletic clades, each with > or =97.9% within-clade 16S rRNA gene sequence identity, were identified. Clade community similarity positively correlated with lake environmental similarity but not with geographic distance, implying that the lakes represent a single biotic region containing environmental filters for communities that have similar compositions. Phylogenetically disparate clades within the acI lineage were most abundant at the regional scale, and local communities were comprised of more closely related clades. Lake pH was a strong predictor of the community composition, but only when lakes with a pH below 6 were included in the data set. In the remaining lakes (pH above 6) biogeographic patterns in the landscape were instead a predictor of the observed acI community structure. The nonrandom distribution of the newly defined acI clades suggests potential ecophysiological differences between the clades, with acI clades AI, BII, and BIII preferring acidic lakes and acI clades AII, AVI, and BI preferring more alkaline lakes.     

Phylogenetic Ecology of the Freshwater Actinobacteria acI Lineage

The acI lineage of freshwater Actinobacteria is a cosmopolitan and often numerically dominant member of lake bacterial communities. We conducted a survey of acI 16S rRNA genes and 16S-23S rRNA internal transcribed spacer regions from 18 Wisconsin lakes and used standard nonphylogenetic and phylogenetic statistical approaches to investigate the factors that determine acI community composition at the local scale (within lakes) and at the regional scale (across lakes). Phylogenetic reconstruction of 434 acI 16S rRNA genes revealed a well-defined and highly resolved phylogeny. Eleven previously unrecognized monophyletic clades, each with ≥97.9% within-clade 16S rRNA gene sequence identity, were identified. Clade community similarity positively correlated with lake environmental similarity but not with geographic distance, implying that the lakes represent a single biotic region containing environmental filters for communities that have similar compositions. Phylogenetically disparate clades within the acI lineage were most abundant at the regional scale, and local communities were comprised of more closely related clades. Lake pH was a strong predictor of the community composition, but only when lakes with a pH below 6 were included in the data set. In the remaining lakes (pH above 6) biogeographic patterns in the landscape were instead a predictor of the observed acI community structure. The nonrandom distribution of the newly defined acI clades suggests potential ecophysiological differences between the clades, with acI clades AI, BII, and BIII preferring acidic lakes and acI clades AII, AVI, and BI preferring more alkaline lakes.     

16S rDNA characterisation of bacterial and archaeal communities during start-up of anaerobic thermophilic digestion of cattle manure

A laboratory-scale continuously stirred anaerobic thermophilic batch digester was inoculated with cattle manure. Bacterial and archaeal communities, as well as digester performances, were analysed during reactor start-up for about 20 days. Polymerase chain reaction-single strand conformation polymorphism (PCR-SSCP) was used for overall detection and for study of the dynamics of microbial populations. Dominant bacteria and archaea 16S rDNAs were sequenced from the sample on day 12. Ten bacteria and 3 archaea OTUs (operational taxonomic units) were identified from the 52 clones sequenced. Sequences corresponding to the dominant bacterial SSCP peak were phylogenetically close to the 16S rDNA sequence of Bacillus thermoterrestris, whereas sequences corresponding to the two dominant archaeal SSCP peaks were phylogenetically close to the 16S rDNA sequence of Methanoculleus thermophilicus and Methanosarcina thermophila.     

Analysis of structure and composition of bacterial core communities in mature drinking water biofilms and bulk water of a citywide network in Germany

The bacterial core communities of bulk water and corresponding biofilms of a more than 20-year-old drinking water network were compared using 16S rRNA single-strand confirmation polymorphism (SSCP) fingerprints based on extracted DNA and RNA. The structure and composition of the bacterial core community in the bulk water was highly similar (>70%) across the city of Braunschweig, Germany, whereas all biofilm samples contained a unique community with no overlapping phylotypes from bulk water. Biofilm samples consisted mainly of Alphaproteobacteria (26% of all phylotypes), Gammaproteobacteria (11%), candidate division TM6 (11%), Chlamydiales (9%), and Betaproteobacteria (9%). The bulk water community consisted primarily of Bacteroidetes (25%), Betaproteobacteria (20%), Actinobacteria (16%), and Alphaproteobacteria (11%). All biofilm communities showed higher relative abundances of single phylotypes and a reduced richness compared to bulk water. Only biofilm communities sampled at nearby sampling points showed similar communities irrespective of support materials. In all of our bulk water studies, the community composition determined from 16S rRNA was completely different from the 16S rRNA gene-based community composition, whereas in biofilms both molecular fractions resulted in community compositions that were similar to each other. We hypothesize that a higher fraction of active bacterial phylotypes and a better protection from oxidative stress in drinking water biofilms are responsible for this higher similarity.     

Molecular community analysis of magnesium-rich bittern brine recovered from a Tunisian solar saltern

The microbial community of a magnesium-rich bittern brine saturated with NaCl (380-400 g/L) from a Tunisian solar saltern was investigated using a molecular approach based on 16S rRNA gene analysis and viability tests. The results revealed the existence of microbial flora. Viability test assessment showed that 46.4% of this flora was viable but not detectable by culturability tests. 16S rRNA genes from 49 bacterial clones and 38 archaeal clones were sequenced and phylogenetically analyzed. Eleven operational taxonomic units (OTUs) determined by the DOTUR program with 97% sequence similarity were generated for Bacteria. These OTUs were affiliated with Bacteroidetes and Gammaproteobacteria. The archaeal community composition exhibited more diversity with 38 clones, resulting in 13 OTUs affiliated with the Euryarchaeota phylum. Diversity measurement showed a more diverse archaeal than bacterial community at the saturated pond.     

16S rRNA Gene Sequence Analysis of Drinking Water Using RNA and DNA Extracts as Targets for Clone Library Development

The bacterial composition of chlorinated drinking water was analyzed using 16S rRNA gene clone libraries derived from DNA extracts of 12 samples and compared to clone libraries previously generated using RNA extracts from the same samples. Phylogenetic analysis of 761 DNA-based clone sequences showed that unclassified bacteria were the most abundant group, representing nearly 62% of all DNA sequences analyzed. Other phylogenetic groups identified included Proteobacteria (20%), Actinobacteria (9%), Cyanobacteria (4%), and Bacteroidetes (2%). The composition of RNA-based libraries (1122 sequences) was similar to the DNA-based libraries with a few notable exceptions: Proteobacteria were more dominant in the RNA clone libraries (i.e., 35% RNA; 20% DNA). Differences in the Proteobacteria composition were also observed; alpha-Proteobacteria was 22 times more abundant in the RNA-based clones while beta-Proteobacteria was eight times more abundant in the DNA libraries. Nearly twice as many DNA operational taxonomic units (OTUs) than RNA OTUs were observed at distance 0.03 (101 DNA; 53 RNA). Twenty-four OTUs were shared between all RNA- and DNA-based libraries (OTU_0.03) representing only 18% of the total OTUs, but 81% (1527/1883) of all sequences. Such differences between clone libraries demonstrate the necessity of generating both RNA- and DNA-derived clone libraries to compare these two different molecular approaches for community analyses.     

A Comparison of Methods for Clustering 16S rRNA Sequences into OTUs

Recent studies of 16S rRNA sequences through next-generation sequencing have revolutionized our understanding of the microbial community composition and structure. One common approach in using these data to explore the genetic diversity in a microbial community is to cluster the 16S rRNA sequences into Operational Taxonomic Units (OTUs) based on sequence similarities. The inferred OTUs can then be used to estimate species, diversity, composition, and richness. Although a number of methods have been developed and commonly used to cluster the sequences into OTUs, relatively little guidance is available on their relative performance and the choice of key parameters for each method. In this study, we conducted a comprehensive evaluation of ten existing OTU inference methods. We found that the appropriate dissimilarity value for defining distinct OTUs is not only related with a specific method but also related with the sample complexity. For data sets with low complexity, all the algorithms need a higher dissimilarity threshold to define OTUs. Some methods, such as, CROP and SLP, are more robust to the specific choice of the threshold than other methods, especially for shorter reads. For high-complexity data sets, hierarchical cluster methods need a more strict dissimilarity threshold to define OTUs because the commonly used dissimilarity threshold of 3% often leads to an under-estimation of the number of OTUs. In general, hierarchical clustering methods perform better at lower dissimilarity thresholds. Our results show that sequence abundance plays an important role in OTU inference. We conclude that care is needed to choose both a threshold for dissimilarity and abundance for OTU inference.