Skip to main content
Fig. 1 | Microbiome

Fig. 1

From: Proportion-based normalizations outperform compositional data transformations in machine learning applications

Fig. 1

Schematic representation of workflow for the creation of data transformations from the raw Dada2 counts tables. This schematic starts with the raw Dada2 counts tables and each step in the workflow (grey arrows with white bubbles) lead to trees (grey blocks) or final datasets (blue boxes) as described in the methods section. Datasets with random node shuffles were recreated at least 3 times

Back to article page

Microbiome

ISSN: 2049-2618

Contact us