Approaches to biomedical knowledge
Ontologies and Large Language Models
This course is held as a part of the Master Curriculum in Bioinformatics of the Free University Berlin.
Lecture Schedule
- Lecture 1: Ontologies: OBO, RDF, RDFS, and OWL
- Lecture 2: Gene Ontology (GO) and Overrepresentation Analysis
- Lecture 3: Bayesian algorithms for GO: Model-based gene-set analysis
- Lecture 4: Human Phenotype Ontology: Semantic similarity algorithms
- Lecture 5: From Logistic Regression to Neural Foundations
- Lecture 6: Neural networks
- Lecture 7: RNNs and LSTMs
- Lecture 8: Embeddings and Tokens
- Lecture 9: Transformers
- Lecture 10: Transformers (continued)
- Lecture 11: BERT and Sentence BERT
- Lecture 12: Reinforcement learning and LLMs
- Lecture 13: RAG, RAG-HPO, SPIRES
- Lecture 14: Applications of LLMs in Medicine: Problems and opportunities
Homework
- Each lecture will be accompanied by a practical session to go over background material and review the homework.
Examples
- This course does not intend to teach practical coding skills, but in some cases it can be instructional to study Python implementations of the algorithms presented in the lectures.
- The following notebooks present simplified versions of algorithms and approaches and are not intended to be full implementations.
background
- Additional explanations of certain topics for the practical sessions