Strategy for microbiome analysis using 16S rRNA gene sequence analysis on the Illumina sequencing platform

Abstract

Understanding the identity and changes of organisms in the urogenital and other microbiomes of the human body may be key to discovering causes and new treatments of many ailments, such as vaginosis. High-throughput sequencing technologies have recently enabled discovery of the great diversity of the human microbiome. The cost per base of many of these sequencing platforms remains high (thousands of dollars per sample); however, the Illumina Genome Analyzer (IGA) is estimated to have a cost per base less than one-fifth of its nearest competitor. The main disadvantage of the IGA for sequencing PCR-amplified 16S rRNA genes is that the maximum read-length of the IGA is only 100 bases; whereas, at least 300 bases are needed to obtain phylogenetically informative data down to the genus and species level. In this paper we describe and conduct a pilot test of a multiplex sequencing strategy suitable for achieving total reads of > 300 bases per extracted DNA molecule on the IGA. Results show that all proposed primers produce products of the expected size and that correct sequences can be obtained, with all proposed forward primers. Various bioinformatic optimization of the Illumina Bustard analysis pipeline proved necessary to extract the correct sequence from IGA image data, and these modifications of the data files indicate that further optimization of the analysis pipeline may improve the quality rankings of the data and enable more sequence to be correctly analyzed. The successful application of this method could result in an unprecedentedly deep description (800,000 taxonomic identifications per sample) of the urogenital and other microbiomes in a large number of samples at a reasonable cost per sample.

Keywords:

• Desulfovibrio vulgaris
• Escherichia coli
• metagenomics
• ribosomal RNA
• Streptococcus pyogenes

Acknowledgments

Fecal samples were obtained from community volunteers by Dr. Ikuko Kato under a research protocol that was approved by Wayne State University and the VA Medical Center Human Investigation Committees. Signed informed consent was obtained from each study participant. The research was supported by internal research stimulation funds provided by the Department of Physiology and Wayne State University. We thank the personnel of Wayne State's Applied Genomics Technology Center for technical assistance in designing these experiments and analyzing the image data.

Declaration of Interest: The authors alone are responsible for the content and writing of the paper and report no conflicts of interest.