Recently, Springer Nature and MassBio hosted a summit at the MassBio Hub in Cambridge, Massachusetts focused on text and data mining, featuring a keynote interview delivered by Mark Davies. In the interview, Mark discusses some of the steps BenevolentAI takes to acquire, process and integrate a wide variety of data sets and types in our machine learning models to improve our drug discovery capabilities, particularly in identifying novel targets for diseases with high unmet medical needs. He also highlights the importance of the scientific literature as a data source, and how our teams extract relationships from such data using natural language processing.
Mark has a background in molecular genetics and bioinformatics and over 15 years of experience working on biomedical data representation, data analysis and application development. In 2001, he joined the London based biotechnology company Inpharmatica, where he was initially working on mining the output of the Human Genome Projects and eventually moved on to building Chemogenomics systems used by pharmaceutical companies, such as Bayer. Mark moved to the European Bioinformatics Institute (EMBL-EBI) as one of the founding members and technical lead for the ChEMBL resource - the largest open-source SAR database. Mark was also responsible for the successful transition of the SureChEMBL chemical patent system from Digital Science to the EMBL-EBI.