Fig. 1: Visualization of exemplary one-dimensional and two-dimensional datasets.
From: Data splitting to avoid information leakage with DataSAIL

The symbol “Y” indicates the presence of a measurement, while the phylogenetic trees next to the matrix visualizations illustrate similarities between samples. The figure showcases all splitting tasks and their interrelations. Samples assigned to training are highlighted with a blue background, validation samples are in yellow, and test samples are marked in red. Unassignable tiles are left white. Created in BioRender. Joeres, R. (2025) https://BioRender.com/w47j283.