Heatmap Analysis

Jump to Section:

• View Settings & Filters
• Setting Initial Filters
• Virus Vs. Bacteria Analysis

You can manipulate the data in the heatmap by setting up multiple filters and adjusting other view settings. We strongly recommend that you start by applying a set of highly restrictive filters to your heatmap to reduce the initial complexity/noise that you see. 

IMPORTANT NOTE: Only taxa that have at least 5 reads in NT or NR will be included in a heatmap. All other filters will be applied on top of this initial default filter.

---

View Settings & Filters

You can adjust the heatmap view settings to adjust the level of taxonomic resolution used for taxa (i.e., genus vs species), the way taxa and samples are sorted, and the metric used to calculate distances displayed in the heatmap using a color scale, including Z Score, reads (r), or reads per million (rPM). Additionally, you can use filters to help you focus on relevant taxa (e.g., by filtering out low abundance taxa). Note that every time you add a filter, the heatmap is recalculated to meet the specified filter criteria. Therefore, you may notice new taxa after changing or adding filters.

View Settings

• Taxon Level: Changes the granularity of the taxon on the heatmap. You can choose either the species level or genus level.
• Sort Taxa: Changes the way the taxa are sorted. By default, we cluster taxa using average-linkage hierarchical clustering based on Euclidean distances between rows. When you have selected species as your taxon level you can also sort taxa by genus.
• Sort Samples: Changes the way that samples are sorted. By default, we cluster taxa using average-linkage hierarchical clustering. You can also sort samples alphabetically.
• Metric: The chosen metric dictates the color value displayed in each heatmap square.
• Background: If you chose Z-score for Metric, the heatmap will use your background model selection here to calculate that value.
• Scale: View the heatmap using a log or linear scale.

Filters

• Categories (Taxon group): Lets you focus on a particular group of taxa (e.g., bacteria, viruses, eukaryotes).
• Threshold Filters: These filters work just like the filters on the single sample report page. You can add as many threshold filters as you want. This will help reduce the complexity of the heatmap. 
• Read Specificity: Select “Species Only” to hide reads that were not defined at the species level.
• Taxa Per Sample: Choose how many taxa per sample to include in the heatmap. Increasing this value will increase heatmap load time.
• Pathogen Tag (Known Pathogens Only): Select this filter to only show known pathogenic taxa in the heatmap. Taxa are tagged as "Known Pathogens" based on the CZ ID pathogen list. Note that this list is not fully comprehensive and does not distinguish pathogenic strains or subtypes from non-pathogenic ones in many cases. Therefore, the pathogen list should be viewed as a starting point to quickly identify potential pathogenic organisms in samples.

---

Setting Initial Filters

To reduce the initial complexity of the heatmap we suggest using the following heatmap settings:

1. Metric: NT rPM
   1. This will adjust the metric plotted on the heatmap, such that higher rPM values appear darker.

2. Taxon Level: Genus
   1. Evaluating the genus level helps you zoom out and see high level trends before diving into species

We recommend beginning with a highly restrictive set of filters to get a high level view of the data.

1. NT rPM >= 10
2. NT L >= 50

If you are looking at samples from the Feline Gingivitis study, your heatmap should look like the one below after applying the suggested view settings and filters.

This heatmap gives a broad picture of the landscape of taxa across your samples.

To view more information about a taxa and a sample hover over the square in the heatmap.

Clicking on the highlighted square will will open the Sample Report page in a new tab.

To remove taxa from the heatmap, hover over the taxon name. Click on the small blue x that  appears on the left-hand side of the screen.

---

Virus Vs. Bacteria Analysis

Sometimes it is useful to separate the analysis of bacteria from that of viruses. Let’s first focus on just viruses. Adjust the Taxon Categories filter = Viruses. Then, since this will reduce the total number of microbes in the heatmap, let’s look more specifically at the species level by adjusting Taxon Level = Species. The resulting heatmap should look like the one below.