I am one of the founders and developers of the Human Phenotype Ontology (HPO). I have been working a lot with different ontologies, starting from obo-formatted ontologies such as Gene Ontology and HPO. These and other ontologies have then been shifted into very expressive OWL2 ontologies, which made me start to dive into the semantic web and description logics world. Read more about my ontology-projects.
A lot of biological data can be represented in graph structures, where the nodes correspond do biological entities and edges between them indicate a relation between two entities. During my PhD and afterwards I have worked a lot with complex networks representing phenotypes, proteins, and disease entities. Read more about my network-related work here.
Machine learning applied to biological problems is also one of my favorite topics. I have used SVMs for classifying disease genes and have worked on statistical learning approaches for ontological similarity calculation. You can read more about this here.