The research and development sector of pharmaceutical companies represents a very interesting case for big data. In the last 50 years this industry produced a formidable amount of information about the chemical compounds they synthesised and assayed. These data integrate physicochemical properties, in-vitro and in-vivo experimental results and stock availability which often cover a size of millions of compounds. Similar effort has been carried out by the public through ChEMBL database. On top of this information, genomics revolution is also providing further challenges and opportunities to be exploited in this realm.
Without assuming a deep chemistry knowledge, I’d like to give an overview of the traditional concepts and tools used in chemical information R&D and the typical tasks that are performed with this data. Additionally I will mention few interesting cases that show how this vast amount of data can be put to fruition to discover new unsuspected relations.

Davide Branduardi received his Master in Material Science in Milan and later did his PhD in Michele Parrinello's research group at ETHZ. Since July 2015 he is a senior applications scientist for Schrodinger Inc.

His main research focus is on free energy estimates of chemical reactions and allosteric transitions via molecular simulations. In particular developing and employing path based methods (Path Collective Variables approaches and String Methods) for optimizing reaction paths.