New SciDAP Workflow: 16S Metagenomic analysis with QIIME2

16S analysis targets the 16S ribosomal RNA gene, a molecular beacon that helps to discover the identities and relationships of microorganisms residing in environments as diverse as the human gut, soil, oceans, and beyond. The analysis of the 16S rRNA gene in a single sample or among groups of samples has emerged as a potent tool, offering unprecedented insight into the hidden diversity and dynamics of microbial communities.

Example of mixed oral bacteria (https://irp.nih.gov/our-research/accelerating-science/zooming-in-on-the-microbiome).

Background

The 16S rRNA gene is a fundamental genetic marker found in the genomes of all bacteria and archaea. This gene evolves at a relatively steady rate, making it an invaluable tool for taxonomic classification. By targeting and sequencing this gene, researchers can identify and differentiate between various microbial taxa, which can then be leveraged to unveil the web of interactions that define a community.

The journey of a 16S sequencing begins with the collection of environmental samples. These could originate anywhere: from rainforest soil samples to the microbial inhabitants of our own gut. Once collected, DNA is extracted and the 16S rRNA gene is PCR (polymerase chain reaction) amplified. The resulting genetic material is then subjected to high-throughput sequencing, producing millions of reads that represent the genetic diversity within the sample.

The primary goals of 16S analysis are multifaceted. First and foremost, it seeks to unravel the taxonomic composition of microbial communities, revealing the identity of the countless microbial taxa present within a given sample. This taxonomic census paves the way for understanding community structure, ecological interactions, and the factors that shape microbial diversity. Beyond taxonomy, 16S analysis aims to resolve the relative abundances of these microbes within a sample, shedding light on their roles and contributions within their ecosystems. Moreover, it serves as a powerful tool for investigating shifts in community composition across diverse conditions, be it in response to environmental changes, disease states, or interventions. Ultimately, 16S analysis empowers researchers to explore the hidden microbial worlds that influence everything from human health to environmental sustainability.

Molecular structure of the 30S (small) subunit (SSU) of ribosome from Thermus thermophilus. Proteins are shown in blue and the single RNA strand in orange. 16S ribosomal RNA (or 16S rRNA) is the RNA component of the 30S subunit of a prokaryotic ribosome (SSU rRNA). [1]

Approaches for Studying Microbial Communities

Studying microbiomes through 16S rRNA analysis involves a powerful suite of molecular techniques. After collection of environmental or clinical samples, total DNA is extracted. Next, the 16S rRNA gene is selectively amplified using PCR, and the resulting gene fragments are then subjected to high-throughput sequencing, generating millions of microbial DNA sequences. These sequences are then processed through bioinformatics pipelines, where they are clustered into operational taxonomic units (OTUs) or amplicon sequence variants (ASVs), allowing for taxonomic identification and abundance estimation. This comprehensive approach enables researchers to not only catalog the microbial diversity within a sample but also explore community dynamics, ecological relationships, and its functional potential.

16S Metagenomic Analysis with QIIME 2

QIIME 2 logo.

QIIME 2 is a powerful, extensible, and decentralized microbiome analysis package with a focus on data and analysis transparency [2]. It enables researchers to start an analysis with raw DNA sequence data and finish with publication-quality figures and statistical results. The pipeline begins with importing your 16S sequencing data. Once imported, quality control steps are performed including demultiplexing (if necessary), trimming low-quality reads, and filtering out any chimeric sequences.

Base quality plot from QIIME 2 [3].

Next, taxonomic classification is carried out using a pre-trained classifier compatible with your 16S region (e.g., Greengenes, SILVA). Then, feature tables and representative sequences are generated from which alpha and beta diversity metrics are calculated. Alpha diversity measures diversity within a single sample, while beta diversity measures differences between samples. These metrics provide insights into the richness and evenness of microbial communities and their similarities or dissimilarities across samples.

Alpha rarefaction plot from QIIME 2 [3].

If a researcher wants to identify taxa that are differentially abundant between groups or conditions, they can perform differential abundance analysis using ANCOM or Gneiss. These tools help identify taxa that show significant abundance differences while considering the compositional nature of 16S data. Finally, visualization of results using various plots, such as bar charts, heatmaps, and PCoA plots help researchers interpret the results in the context of their research questions, considering factors like sample metadata and the ecological context of the study.

Taxonomic classification bar plot from QIIME 2 [3].

The SciDAP Solution for 16S Metagenomic Analysis

SciDAP offers the QIIME 2 pipeline described above for single sample analysis, and an aggregated workflow that combines multiple samples for an comparative analysis. The workflow starts with importing all sample read data that is then summarized in both table and graph formats. Next, denoising occurs which will detect and correct (where possible) Illumina amplicon sequence data. This process will additionally filter out any phiX reads (control sequence commonly used in Illumina sequencing) and identify and filter chimeric sequences. A phylogenetic tree is generated for diversity analyses and rarefaction processing and plotting. Rarefaction plots show the relationship between sampling depth and alpha diversity which is expected to be positively correlated. Since increased sequencing depth may allow detection of lower abundance species, comparison of samples sequenced to different depths can result in misleading analyses. Researchers typically will choose a sampling depth for normalization prior to differential abundance analysis by setting it at a value where alpha diversity begins to plateau. If some samples have a depth lower than this value, they will be dropped from downstream analyses. Taxonomy classification of amplicons is performed using a Naive Bayes classifier trained on the Greengenes2 database “gg_2022_10_backbone_full_length.nb.qza”.

After the analysis of individual samples is complete, comparative analysis will aggregate these data for all samples and perform statistical comparisons between sample groups defined in the user-provided sample metadata file. Principal coordinates analysis (PCoA) may be performed using the unweighted unifrac and bray curtis methods if an appropriate metadata variable is provided (must be integer values). 3D plots showing the relationship between principal coordinates and the metadata variable of interest are then produced and available for viewing with the Emperor visualization tool. If the inputs for sampling depth and metadata header for grouped differential analysis are provided, differential abundance analysis will be performed using ANCOM at the family, genus, and species taxonomic levels. Volcano plots between the specified group will be produced for each taxonomic level and a list of significantly significantly differentially abundant taxa is also made available. For a given taxon, ANCOM accounts for the compositionality of microbiome data by calculating a W statistic, which represents the number additive log ratio (ALR) transformed models where the taxon is differentially abundant with regard to the variable (or group) of interest [4]. Finally, an unsupervised hierarchical clustering heatmap (Gneiss) is produced that may aid in identifying patterns of taxonomic co-abundance, differential abundance between groups, sample similarity, and other ecological insights. SciDAP 16S metagenomic workflow outputs include an overview of various input parameters and all above images from QIIME 2.

To start your 16S metagenomic analysis in SciDAP, please follow this step-by-step quick-guide:

1. After registering your account and beginning a free trial, create your laboratory at the prompt when logging in.
2. Next, create a project where your data will be stored and processed. When creating a project, you’ll need to name it, include any subtitles or extra information about your experiment in the abstract section if desired, and select all applicable workflows. For 16S Metagenomic analysis, use the following workflows: “16S metagenomic paired-end with QIIME2” and “16S metagenomic paired-end QIIME2 Analysis (differential abundance)”, for individual samples and comparative analyses respectively.
3. Save the new Project, then click “ADD SAMPLE” to begin inputting your samples with their metadata and sequence read data. Enter a name for your sample, the environment from which the sample came, and a catalog number if available. Input your FASTQ files using the File Manager, FTP connection, or SRA accession number if publicly available. Finally, set your read trimming and truncation values for your forward and reverse reads (there are tooltips next to each input to help you figure out what values are appropriate for your samples).
4. Select “Save sample” and your sample will be queued for processing. Upon completion there will be a list of sample inputs on the Overview tab, along with up to three different links out to the QIIME 2 visualization page that will visualize your data: FASTQ summary file, alpha rarefaction curve, and taxonomic abundance barplot. All intermediate and final output files are also made available for download.
5. After all samples in your 16S experiment have been processed, go to the Analysis tab and click “ADD ANALYSIS” to aggregate your samples and perform principal coordinate and differential abundance analysis. Name your analysis, add your metadata sample sheet (please see our tutorial video on YouTube for an example), PCoA axis label, sampling depth for normalization, group name for differential analysis, and finally select all the samples you want to aggregate. Again, all inputs have a tooltip to help you choose the appropriate input values.
6. Once processed, you will have several links available on the Overview tab of the Analysis sample, all of which are described above, and in our YouTube tutorial.

Watch on Youtube: https://youtu.be/_i_gAxoH64s

Now biologists can easily perform 16S metagenomic analysis without learning to code, running command line tools, or running commands in R. With the SciDAP 16S Metagenomic workflow, you can move beyond the command line to easily generate microbial community profiles and identify differences between them. Try it for yourself with a FREE trial.

References

1. Hall, M., Beiko, R.G. (2018). 16S rRNA Gene Analysis with QIIME2. In: Beiko, R., Hsiao, W., Parkinson, J. (eds) Microbiome Analysis. Methods in Molecular Biology, vol 1849. Humana Press, New York, NY. https://doi.org/10.1007/978-1-4939-8728-3_8
2. Bolyen E, et al. 2019. Reproducible, interactive, scalable and extensible microbiome data science using QIIME 2. Nature Biotechnology 37: 852–857. https://doi.org/10.1038/s41587-019-0209-9
3. Qiime2. QIIME2 — MAVERIC Informatics documentation. (n.d.). https://maveric-informatics.readthedocs.io/en/latest/QIIME2.html
4. ANCOM Tutorial. Huang Lin, NIEHS, Research Triangle Park, NC 27709, USA. July 06, 2023. https://bioconductor.org/packages/release/bioc/vignettes/ANCOMBC/inst/doc/ANCOM.html