CheTo (ChemicalTopic) allows to apply topic modeling, a method developed in the text-mining field, to chemical data. Please see our recent publication for detailed information:
Schneider, N.; Fechner, N.; Landrum, G. A.; Stiefl, N. Chemical Topic Modeling: Exploring Molecular Data Sets Using a Common Text-Mining Approach. J. Chem. Inf. Model. 2017, http://pubs.acs.org/doi/10.1021/acs.jcim.7b00249
The supplementary of the paper contains exemplary data sets extracted from the ChEMBL database and Jupyter notebooks to run the experiments described in the paper.
An interactive web page showing an exemplary topic model of data set A from our paper can be found here http://www.t5informatics.com/Papers/InteractiveTopicModelDatasetA.html
Installation
To install CheTo using Conda, simply run:
conda install -c rdkit cheto
Further reading
Using CheTo in KNIME: http://rdkit.blogspot.ch/2017/08/chemical-topic-modeling-with-rdkit-and.html
After publication of our article we were made aware that applying topic modeling to chemical data was also suggested by Rajarshi Guha in 2012 in his blog (http://blog.rguha.net/?p=997).