Context:
Networks have long been popular means for modeling complex biological systems where they model interactions between biological entities and their effects on various biological systems. Biomedical networks can also be converted into so called knowledge graphs, formal representations of entities and semantic relationships between them.
Consequently, knowledge graphs can be used for developing relational machine learning models which are known to provide highly scalable and accurate predictions of associations between biomedical entities, such as drugs, proteins or diseases. And a prime example of a high-impact application of the predictive models powered by knowledge graphs is computational drug re-purposing (i.e., finding new and potentially groundbreaking uses for drugs that are already approved for other diseases).

Goals:
- Review state of the art on (biomedical) knowledge graphs.
- Develop a sufficiently representative knowledge graph covering approved drugs, their targets and diseases (and possibly other relevant entity types).
- Design a new knowledge graph embedding model (or adapt an existing one) for predicting new indications of drugs originally approved for other diseases.
- Apply one or more interpretable ML and XAI techniques to the developed drug repurposing model to explain its predictions.
- Provide a preliminary validation of the predictions and their explanations, either by using existing benchmarks, or developing a new one, or by combining literature search with feedback of independent experts.
- Write up the results in a thesis form.

Requirements:
- Keen interest in the topic.
- Ability to read and comprehend scientific publications in English.
- Basic knowledge of computational biology, bioinformatics or medicine is a plus, but certainly not a must - the interest and willingness to learn are the only essential requirements, really.
- While the thesis can be written and defended in Czech, its elaboration and presentation in English will be supported enthusiastically.
- Monthly (or more frequent, if needed) progress review meetings with the supervisor will be expected.
- The student(s) will also be expected to develop and document any related code using the FI MU Gitlab platform, and (if applicable) re-use and interact with other related projects there.

Notes:
- This topic can be picked by more than one student, however, their works must be complementary in terms of the implemented methods and use cases covered.
- Should they be interested, the student may tap into the resources and informal feedback of a broader international group of biomedical AI researchers and practitioners based in the Czech Republic, Ireland and elsewhere.
- The student will also be very welcome to get involved in related inter-disciplinary research projects involving the Faculties of Informatics and Science, and Masaryk Memorical Cancer Institute.