Tutorial Extension

Extension

Input preparation

Prior to following this tutorial, please download input files here

You should have following files:

1. PCoA files (weighted/unweighted UniFrac, Bray-Curtis, Jaccard) (.qza)
2. Feature table (labelled as merged_table.qza)
3. Taxonomic classification result (labelled as taxonomy.qza)
4. Metadata file (.tsv)
5. Environmental metadata file (.tsv)

Custom PCoA plot

PCoA is used to project samples into multidimensional space while maximally preserving the original dissimilarity relationship between the samples.

While QIIME2 offers a way to visusalize ordination via emperor plot, you may use AXIOME3 to generate more customizable PCoA plot.

Input(s)

1. QIIME2 archived PCoA artifact (.qza) (output from "Analysis" module) AXIOME3 generates four different types of PCoA ordinations,
   1. Unweighted UniFrac (named as 'unweighted_unifrac_pcoa.qza')
   2. Weighted UniFrac (named as 'weighted_unifrac_pcoa.qza')
   3. Bray-Curtis (named as 'bray_curtis_pcoa.qza')
   4. Jaccard (named as 'jaccard_pcoa.qza')
2. Metadata file (refer to the sample metadata file) (it will be used to overlay metadata groups on PCoA ordination)

Steps

For the tutorial, we will be using bray_curtis_pcoa.qza that was generated in the Analysis module, and the sample_metadata.tsv.

1. Upload input files

2. Choose options

   1. Fill variable: First metadata column to visualize. It will be represented as different colours. Type metadata column name you want to visualize (case sensitive). We will use Year for this option.
   2. Fill variable data type: Fill variable data type; numeric or categorical?
   3. Shape variable (optional): Second metadata column to visualize. It will be represented as different shapes.
   4. Colour set: Colour pallette to use to visualize the Fill variable. We will use Paired (default option).
   5. You may furthur customize point and plot related options in the Points Config and Plot Layout Config tabs. (we will skip this for tutorial, but feel free to play around with it)

3. Start analysis

When it's done running it should look like the following

Triplot

Bubble plot