Abstract
The challenge of analyzing high-dimensional data affects many scientific disciplines, from pharmacology to chemistry and biology. Traditional dimensionality reduction methods often oversimplify data, making it difficult to interpret individual points. This distortion can complicate the visualization of mutual distances between data points in the reduced space. Graphs provide an effective framework for representing objects and their relationships. One of their possible use is visualizing similarity patterns in tabular datasets. Here we introduce TaGra, an off-the-shelf package designed to generate a graph of similarity relations from tabular data. TaGra enables the visualization of datasets in 2D space, identification of typical data points and outliers, and assessment of the separation between items with different target variables. We describe TaGra's functionality, options and setup. The software including examples, instructions and a guide, is openly available on PyPI at https://pypi.org/project/TaGra/ and on GitHub at https://github.com/davidetorre92/TaGra.