Limits...
Second-generation environmental sequencing unmasks marine metazoan biodiversity.

Fonseca VG, Carvalho GR, Sung W, Johnson HF, Power DM, Neill SP, Packer M, Blaxter ML, Lambshead PJ, Thomas WK, Creer S - Nat Commun (2010)

Bottom Line: We show that remarkable differences in diversity occurred at microgeographical scales and refute currently accepted ecological and taxonomic paradigms of meiofaunal identity, rank abundance and concomitant understanding of trophic dynamics.However, the slope of taxon rarefaction curves for many phyla remains incomplete, suggesting that the true alpha diversity is likely to exceed current perceptions.The approaches provide a rapid, objective and cost-effective taxonomic framework for exploring links between ecosystem structure and function of all hitherto intractable, but ecologically important, communities.

View Article: PubMed Central - PubMed

Affiliation: Molecular Ecology and Fisheries Genetics Laboratory, School of Biological Sciences, Environment Centre Wales, Bangor University, Deiniol Road, Gwynedd LL57 2UW, UK.

ABSTRACT
Biodiversity is of crucial importance for ecosystem functioning, sustainability and resilience, but the magnitude and organization of marine diversity at a range of spatial and taxonomic scales are undefined. In this paper, we use second-generation sequencing to unmask putatively diverse marine metazoan biodiversity in a Scottish temperate benthic ecosystem. We show that remarkable differences in diversity occurred at microgeographical scales and refute currently accepted ecological and taxonomic paradigms of meiofaunal identity, rank abundance and concomitant understanding of trophic dynamics. Richness estimates from the current benchmarked Operational Clustering of Taxonomic Units from Parallel UltraSequencing analyses are broadly aligned with those derived from morphological assessments. However, the slope of taxon rarefaction curves for many phyla remains incomplete, suggesting that the true alpha diversity is likely to exceed current perceptions. The approaches provide a rapid, objective and cost-effective taxonomic framework for exploring links between ecosystem structure and function of all hitherto intractable, but ecologically important, communities.

Show MeSH
Lineage-through-time plots for OCTUPUS and ESPRIT.Mean number of OTUs plotted against each percentage identity cut-off (90–99% similarity) generated using 5,000 subsampled sequences (>199 bases in length) from three independent sample sites (Prestwick 2, 7 and Littlehampton 1) using OCTUPUS (blue squares) and ESPRIT's HCluster13 (red circles) OTU clustering. Values are given as average+s.d. (n=9).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2963828&req=5

f1: Lineage-through-time plots for OCTUPUS and ESPRIT.Mean number of OTUs plotted against each percentage identity cut-off (90–99% similarity) generated using 5,000 subsampled sequences (>199 bases in length) from three independent sample sites (Prestwick 2, 7 and Littlehampton 1) using OCTUPUS (blue squares) and ESPRIT's HCluster13 (red circles) OTU clustering. Values are given as average+s.d. (n=9).

Mentions: nSSU rRNA amplicons were generated from eight benthic samples from the low-tide zone of an estuarine beach near Prestwick on the West coast of Scotland, and from one sample from a beach in Littlehampton in the South of England. The amplicons were processed for sequencing on a Roche 454 FLX platform generating a total of 353,896 sequences that were quality filtered to 305,702 for downstream analysis (Table 1). When performing metagenetic assessments of taxon richness, it is important to cluster taxonomic units in a biologically meaningful manner, as even slight differences in similarity cut-offs and using different algorithms can result in significantly different estimates of richness. The taxon clustering comparisons between ESPRIT13 and OCTUPUS (the Operational Clustering of Taxonomic Units from Parallel UltraSequencing, supplementary software available at http://octupus.sourceforge.net/) (Fig. 1) on the subsampled beach pyrosequencing data show that ESPRIT overestimates Operational Taxonomic Units (OTUs) richness compared with OCTUPUS (between 1.1x–4.4x, over the 90–99% cut-off range). Phylogenetic14 and barcoding15 studies based on the analysis of chain-termination nSSU gene sequences suggest that intraspecific divergence at least in nematode species is low (1–2%). However, the true level of intragenomic and intraspecific variation among rRNA gene repeats is largely unknown, and genome-wide analyses reveal a dominant set of conserved sequences accompanied by rare variant sequences16. Our benchmarking exercise, performed against a reference control data set comprising 41 species17, revealed that the 96% similarity OCTUPUS clustering algorithm with accompanying chimera screening estimated taxon richness that was most closely aligned with species richness. At 96% similarity, OCTUPUS resulted in 37 operational clustered taxonomic units (OCTUs), whereas at 97%, OCTUPUS resulted in 51 OCTUs from the control metagenetic analysis17. Thus, although we may be underestimating richness by at least 10%, we have opted for a more conservative approach of setting a 96% identity OCTU cut-off for all subsequent numerical comparisons. At this cut-off, an OCTU is likely to (at worst) correspond to a group of related species. Following the 96% similarity OCTUPUS clustering strategy, the total number of putatively non-chimeric tag reads and OCTUs was 217,221 and 428, respectively. Before chimera screening, 1013 OCTUs were clustered from the initial quality-screened 305,702 reads.


Second-generation environmental sequencing unmasks marine metazoan biodiversity.

Fonseca VG, Carvalho GR, Sung W, Johnson HF, Power DM, Neill SP, Packer M, Blaxter ML, Lambshead PJ, Thomas WK, Creer S - Nat Commun (2010)

Lineage-through-time plots for OCTUPUS and ESPRIT.Mean number of OTUs plotted against each percentage identity cut-off (90–99% similarity) generated using 5,000 subsampled sequences (>199 bases in length) from three independent sample sites (Prestwick 2, 7 and Littlehampton 1) using OCTUPUS (blue squares) and ESPRIT's HCluster13 (red circles) OTU clustering. Values are given as average+s.d. (n=9).
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2963828&req=5

f1: Lineage-through-time plots for OCTUPUS and ESPRIT.Mean number of OTUs plotted against each percentage identity cut-off (90–99% similarity) generated using 5,000 subsampled sequences (>199 bases in length) from three independent sample sites (Prestwick 2, 7 and Littlehampton 1) using OCTUPUS (blue squares) and ESPRIT's HCluster13 (red circles) OTU clustering. Values are given as average+s.d. (n=9).
Mentions: nSSU rRNA amplicons were generated from eight benthic samples from the low-tide zone of an estuarine beach near Prestwick on the West coast of Scotland, and from one sample from a beach in Littlehampton in the South of England. The amplicons were processed for sequencing on a Roche 454 FLX platform generating a total of 353,896 sequences that were quality filtered to 305,702 for downstream analysis (Table 1). When performing metagenetic assessments of taxon richness, it is important to cluster taxonomic units in a biologically meaningful manner, as even slight differences in similarity cut-offs and using different algorithms can result in significantly different estimates of richness. The taxon clustering comparisons between ESPRIT13 and OCTUPUS (the Operational Clustering of Taxonomic Units from Parallel UltraSequencing, supplementary software available at http://octupus.sourceforge.net/) (Fig. 1) on the subsampled beach pyrosequencing data show that ESPRIT overestimates Operational Taxonomic Units (OTUs) richness compared with OCTUPUS (between 1.1x–4.4x, over the 90–99% cut-off range). Phylogenetic14 and barcoding15 studies based on the analysis of chain-termination nSSU gene sequences suggest that intraspecific divergence at least in nematode species is low (1–2%). However, the true level of intragenomic and intraspecific variation among rRNA gene repeats is largely unknown, and genome-wide analyses reveal a dominant set of conserved sequences accompanied by rare variant sequences16. Our benchmarking exercise, performed against a reference control data set comprising 41 species17, revealed that the 96% similarity OCTUPUS clustering algorithm with accompanying chimera screening estimated taxon richness that was most closely aligned with species richness. At 96% similarity, OCTUPUS resulted in 37 operational clustered taxonomic units (OCTUs), whereas at 97%, OCTUPUS resulted in 51 OCTUs from the control metagenetic analysis17. Thus, although we may be underestimating richness by at least 10%, we have opted for a more conservative approach of setting a 96% identity OCTU cut-off for all subsequent numerical comparisons. At this cut-off, an OCTU is likely to (at worst) correspond to a group of related species. Following the 96% similarity OCTUPUS clustering strategy, the total number of putatively non-chimeric tag reads and OCTUs was 217,221 and 428, respectively. Before chimera screening, 1013 OCTUs were clustered from the initial quality-screened 305,702 reads.

Bottom Line: We show that remarkable differences in diversity occurred at microgeographical scales and refute currently accepted ecological and taxonomic paradigms of meiofaunal identity, rank abundance and concomitant understanding of trophic dynamics.However, the slope of taxon rarefaction curves for many phyla remains incomplete, suggesting that the true alpha diversity is likely to exceed current perceptions.The approaches provide a rapid, objective and cost-effective taxonomic framework for exploring links between ecosystem structure and function of all hitherto intractable, but ecologically important, communities.

View Article: PubMed Central - PubMed

Affiliation: Molecular Ecology and Fisheries Genetics Laboratory, School of Biological Sciences, Environment Centre Wales, Bangor University, Deiniol Road, Gwynedd LL57 2UW, UK.

ABSTRACT
Biodiversity is of crucial importance for ecosystem functioning, sustainability and resilience, but the magnitude and organization of marine diversity at a range of spatial and taxonomic scales are undefined. In this paper, we use second-generation sequencing to unmask putatively diverse marine metazoan biodiversity in a Scottish temperate benthic ecosystem. We show that remarkable differences in diversity occurred at microgeographical scales and refute currently accepted ecological and taxonomic paradigms of meiofaunal identity, rank abundance and concomitant understanding of trophic dynamics. Richness estimates from the current benchmarked Operational Clustering of Taxonomic Units from Parallel UltraSequencing analyses are broadly aligned with those derived from morphological assessments. However, the slope of taxon rarefaction curves for many phyla remains incomplete, suggesting that the true alpha diversity is likely to exceed current perceptions. The approaches provide a rapid, objective and cost-effective taxonomic framework for exploring links between ecosystem structure and function of all hitherto intractable, but ecologically important, communities.

Show MeSH