The life sciences have invested significant resources in the development and application of semantic technologies to make research data accessible and interlinked, and to enable the integration and analysis of data. Utilizing the semantics associated with research data in data analysis approaches is often challenging. Now, novel methods are becoming available that combine symbolic methods and statistical methods in Artificial Intelligence. In my talk, I will describe how to apply knowledge graph embeddings for analysis of biological and biomedical data, in particular identification of gene-disease associations and drug targets. I will also show how background knowledge from ontologies can be encoded in a deep neural network model to significantly improve protein function prediction and outperform most state of the art function prediction methods, even when predicting function only from protein sequence.
Robert Hoehndorf is an Assistant Professor in Computer Science at King Abdullah University of Science and Technology in Thuwal. His research focuses on the applications of knowledge representation and reasoning in biology and biomedicine, with a particular emphasis on integrating and analyzing heterogeneous, multimodal data. He is an associate editor for the Journal of Biomedical Semantics, BMC Bioinformatics and editorial board member of the journal Data Science. He published over 80 papers in journals and international conferences.