Poster: Analyzing Tox21 data
Development of robust toxicity and bioactivity models requires investigating diverse chemical space data sets.
We present a new software application to analyze and prepare the datasets of small organic molecules.
Diversity Genie (1) allows a user to easily visualize the groupings of similar molecules on a Sammon’s map – a two dimensional projection which aims to preserve the inter-molecule distances. It also enables to calculate, sort, and filter by various molecular properties, and to interconvert between SD, SMILES, and InChI formats.
We demonstrate our visualization of Tox21 challenge (2) data sets, the analysis of the diversity of molecules and the workflow for data preparation for machine learning modeling.