Their scientists were spending a large amount of time and effort to identify and vet new drug targets due to the fact they had to manually glean insights from many diverse datasets without any way to merge and unify these explorations. They wanted a tool that could create a single interface and elegantly stitch together disparate data in order to increase processivity of ideas, support novel hypotheses, and ultimately result in a quicker and more successful lead candidate pipeline.
We curated a set of 3rd party biological datasets that spanned the breadth of information necessary to understand the relationship between drugs, disease, and genes/proteins. Combining these datasets with the knowledge network extracted using NLP and entity recognition of biomedical abstracts provides a biological network. We designed a UI/UX experience to provide a single location to support the breadth and complexity of their biological investigations ranging from enrichment analyses of pathways and diseases; novel drug target identification; hypothesis generation; and suggestions for future experimentation.