Improvements to in silico predictivity after access to proprietary data
Derek Nexus is built mainly using publicly available data and covers a diverse range of chemical space. However, there are gaps in this chemical space that can be filled by proprietary data. This talk will present two data sharing case studies, Ames test data for mutagenicity and LLNA data for skin sensitisation. Donation of these data sets led to the development of a number of new and/or modified mutagenicity and skin sensitisation alerts which significantly improved the predictivity of Derek Nexus for both the donated proprietary data sets and a public data set.