The nature of the data sources vary and therefore have different requirements to keep them updated. Many remain stable and provide constant updates for every one of our releases—such as ChEMBL (13), COSMIC (18), Reactome (19), ExpressionAtlas (20), Cancer Gene Census (21) or ePMC (22). Some other providers (such as IntOgen) undergo major upgrades that require changes to our data model or considerable manual curation effort to integrate. During 2020, we have also included the curation of COVID-19 evidence, for example the 299 clinical trials with COVID-19 as primary indication and known drug targets annotated by ChEMBL. The most important changes to our evidence throughout the most recent period are shown in Figure 3. Supplementary Figure S1 provides an overview of the number of validated evidence strings from each individual source for each Platform release.
