My role as a data scientist involves integrating vast amounts of medical knowledge, creating intelligent applications powered by knowledge graphs, and enabling other scientists in exploiting knowledge graphs to derive new insights to support drug discovery and development. 

I joined AstraZeneca in 2019 from Babylon Health and immediately felt supported by the company and by my team. They have helped me to settle in quickly and I can feel the impact of my work.

I recently delivered a prototype for a Biological Insights Knowledge Graph comprised of literature derived knowledge, public data sources, and proprietary clinical and pre-clinical multi-layered data including screening, NGS, multi-omics, imaging, CRISPR, clinical and electronic health records. Together with our scientists, our aim is to interrogate this knowledge graph to discover novel drug targets, aid pattern identification defining distinct patient groups and biomarkers, understand drug mechanism of action, and build predictive models for precision medicine.

In my opinion AstraZeneca is one of the world’s most exciting biopharma companies because in recent years it has fully embraced AI across the whole of its operations.

What excites me most about my role is the notion of applying AI in scientifically interesting and challenging settings in order to make a real impact.

David Geleta Data Scientist


Data Scientist


Presented at the International Semantic Web Conference (ISWC)


Obtained PhD in Computer Science (AI), University of Liverpool, Liverpool.


Featured in the Guardian newspaper for a University project

  Featured publications

A novel approach and practical algorithms for ontology integration.

2018, ISWC.

Supporting Digital Healthcare Services Using Semantic Web Technologies.

2018, ISWC2018

Minimal Coverage for Ontology Signatures.

2016, OWL-ED.

An investigation of definability in ontology alignment.

2016, EKAW.

Development of ClickClinica: a novel smartphone application to generate real-time global disease surveillance and clinical practice data, BMC medical informatics and decision making.