A paper entitled “Big Data from Pharmaceutical Patents: A Computational Analysis of Medicinal Chemists’ Bread and Butter” has been published in the Journal of Medicinal Chemistry, authored by NextMove Software Ltd in collaboration with organic chemists at the Novartis Institutes for BioMedical Research (NIBR).
In the paper the authors from NextMove – who develop next-generation chemoinformatics solutions to pharmaceutical industry problems – and NIBR present an investigation of chemical reactions and molecules retrieved from U.S. patents over the past 40 years (1976-2015). As acknowledged in the paper, the study used patent data from PatBase – the global, full text patent database developed by Minesoft in partnership with RWS Group – to determine the set of pharmaceutically relevant patents. A summary of the findings in this paper can be found in this blog post from Derek Lowe.
Thanks to advances in chemical text-mining technology, the authors were able to process vast quantities of patent text and chemical structures from PatBase data for this project. Extracting chemical information from patents is an area of increasing interest for patent information specialists at Minesoft. In 2015 they partnered with NextMove to develop Minesoft Chemical Explorer, using advanced data-mining software to create a chemical patent database searchable by structure or terminology that can link directly to PatBase and other external sources.
The Journal of Medicinal Chemistry article can be accessed here (please note there is a paywall): http://pubs.acs.org/doi/pdf/10.1021/acs.jmedchem.6b00153
More information about Minesoft Chemical Explorer can be found here.