kraut

dendrogram

Plots a hierarchical clustering dendrogram from a wide abundance table or from multiple Kraken/Bracken report files. Distances use the same methods as kraut beta.

Syntax

kraut dendrogram [OPTIONS] INPUT_FILES...

Arguments

Argument	Type	Description
INPUT_FILES...	PATH	Required. One wide abundance table, or two or more Kraken/Bracken report files.

Options

Option	Short	Type	Description
--output	-o	PATH	Output dendrogram plot (.html, .png, .svg, .pdf).
--distance	-d	TEXT	Distance method: braycurtis, aitchison, or jaccard (default: braycurtis).
--clustering	-c	TEXT	Linkage method: ward, average, single, or complete (default: ward).
--rank	-r	TEXT	Taxonomic rank to use when reading report files (default: S).
--abundance-metric		TEXT	Report abundance metric: TOT cumulative counts or LVL taxon-specific counts (default: TOT).
--include-unclassified			Include unclassified reads as a feature.
--pseudocount		FLOAT	Pseudocount added before CLR for Aitchison distance (default: 1).
--min-feature-count		FLOAT	Keep taxa with total abundance at least this value.
--min-samples		INTEGER	Keep taxa detected in at least this many samples (default: 1).
--presence-threshold		FLOAT	Presence cutoff for Jaccard; a feature is present when value is greater than this (default: 0).
--metadata		PATH	Optional sample metadata CSV or TSV file; the first column must match sample names.
--color-by		TEXT	Metadata column used to color sample labels.

Clustering Methods

• ward: treats clustering as a variance minimization problem.
• average: distance between clusters is the average distance between all points in the two clusters.
• single: distance between clusters is the nearest pair of points.
• complete: distance between clusters is the furthest pair of points.

Examples

kraut dendrogram reports/*.tsv \
  --distance braycurtis \
  --clustering ward \
  --output dendrogram.html

kraut dendrogram reports/*.tsv \
  --metadata metadata.tsv \
  --color-by Treatment \
  --output dendrogram.svg

---

← Back to Commands

This site is open source. Improve this page.