Submitted to: Advances in Genome Biology and Technology
Publication Type: Abstract only
Publication Acceptance Date: 12/4/2013
Publication Date: N/A
Citation: Interpretive Summary:
Technical Abstract: The sequence of variable regions along the 16S ribosomal RNA gene is often used to conduct metagenomic surveys of bacterial populations in specific habitats, because of the inter-species variability in these regions and because it is possible to design amplification primers in sections of the gene that are highly conserved across a wide swath of the bacterial kingdom. High throughput sequencing has permitted a more accurate and thorough characterization of these microbiomes, including rarer members of the environment. However, the results can have biases from many sources, including the sampling and DNA extraction methods, specific primers chosen, and variable regions selected for amplification. The latter bias has been due to the length limitations on next-gen platforms, which have generally led to selection of, for example, variable regions 1-3 (v1-v3), or v3-v5. Multiple publications have documented that this selection can have impact on the results, suggesting a more general method would be desirable. We characterize ecological niches associated with beef cattle production, including areas of relatively low bacterial diversity (deep nares or nasopharyngeal) and very high diversity samples (recto-anal mucosa), using long-read SMRT sequencing in circular consensus mode on a Pacific Biosciences RSII instrument. We compare results using the P4/C2 and P5/C3 chemistries, as well as profiles produced by amplicons covering v1-v6, v3-v8, and v1-v8. We also examine depth of sequencing impacts on the profiles and determine a minimal number of cells required to obtain robust profiles, by comparing to shorter read amplicons covering either v1-v3 or v3-v5 sequenced to high depth on a MiSeq platform. We demonstrate that shorter amplicons introduce bias from lack of complete information compared to the v1-v8 amplicon, and that long read technology appears to provide a superior profile of the total microbial content in these environments.