Depicting more accurate pictures of protistan community complexity using pyrosequencing of hypervariable SSU rRNA gene regions

• Published in: Environmental microbiology Vol. 13; no. 2; pp. 340 - 349
• Main Authors: Behnke, Anke, Engel, Matthias, Christen, Richard, Nebel, Markus, Klein, Rolf R, Stoeck, Thorsten
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01-02-2011; Society for Applied Microbiology and Wiley-Blackwell
• Subjects: Bacteria; Bacteria - classification; Bacteria - genetics; Biodiversity; Cluster Analysis; Development Biology; DNA, Bacterial; DNA, Bacterial - genetics; Genes, rRNA; Life Sciences; Sequence Analysis, DNA; Sequence Analysis, DNA - methods
• ISSN: 1462-2912; 1462-2920
• DOI: 10.1111/j.1462-2920.2010.02332.x

Initial environmental pyrosequencing studies suggested highly complex protistan communities with phylotype richness decisively higher than previously estimated. However, recent studies on individual bacteria or artificial bacterial communities evidenced that pyrosequencing errors may skew our view of the true complexity of microbial communities. We pyrosequenced two diversity markers (hypervariable regions V4 and V9 of the small-subunit rDNA) of an intertidal protistan model community, using the Roche GS-FLX and the most recent GS-FLX Titanium sequencing systems. After pyrosequencing 24 reference sequences we obtained up to 2039 unique tags (from 3879 V4 GS-FLX Titanium reads), 77% of which were singletons. Even binning sequences that share 97% similarity still emulated a pseudodiversity exceeding the true complexity of the model community up to three times (V9 GS-FLX). Pyrosequencing error rates were higher for V4 fragments compared with the V9 domain and for the GS-FLX Titanium compared with the GS-FLX system. Furthermore, this experiment revealed that error rates are taxon-specific. As an outcome of this study we suggest a fast and efficient strategy to discriminate pyrosequencing signals from noise in order to more realistically depict the structure of protistan communities using simple tools that are implemented in standard tag data-processing pipelines.