Case Study: Text Mining at Roche pRED
Situation: Roche medicinal chemists were spending too long querying the ever-increasing flood of journal articles, patents, and other sources as they explored compound/target/disease relationships. A typical search might take two weeks, and Roche wanted to see if chemically aware text mining could speed this process, to enable more rapid and better-informed scientific decision-making.
Solution: Roche developed the Artemis system based on the Linguamatics I2E text-mining solution, augmented with ChemAxon’s chemical annotation and name-to-structure tools, to extract and organize compound/target/disease relationships.
Success: Medicinal chemists at Roche use the Artemis system to ask compound-, target-, or disease-centric questions of external and internal data sources. They can explore the augmented hits sets to see, for example, which targets play a role in a specific disease; filter out unwanted compounds; and look at chemical diversity. Typical search times have been cut from around two weeks to a few hours, representing a 40-fold increase in speed.