323103 VU Data science approaches for advancing drug discovery - MPS5 (2020S)

Data science has become an important field in drug discovery and development nowadays. It offers a level of understanding of health, disease and treatment on a scale never before imagined, i.e. it can help researchers to find new drugs or re-use old drugs for new indications. In this lecture and seminar we will talk about existing data sources in the open domain, database schemes/structures, statistics, chemical data & cheminformatics approaches in drug discovery etc. We will further introduce you into the data management/manipulation tools Knime and R and give some more specialized seminars on the uses of KNIME and R in some of the expanding fields in data science (such as machine learning).

Drug discovery and development: from old paradigms to rational approaches:

- VO 1: Traditional drug discovery paradigms and rational drug discovery

Data-driven drug discovery: the holistic view:

- VO 2: Introduction to data-driven drug discovery and translational medicine
- VO 3: Statistics for data sciences: Distributions, correlations, hypothesis testing
- VO 4: Basics of statistical learning
- VO 5: Chemical Data: chemical structure representations, chemical descriptors
- VO 6: Cheminformatics approaches in drug discovery
- VO 7: Data sources, data integration: ChEMBL, OpenPHACTS
- VO 8: Database scheme/structures, Data querying

Predictive Modelling with R/KNIME (Practical part):

- UE 1: QSAR and machine learning in KNIME
- UE 2: Selectivity profiling in KNIME
- UE 3: Applied Cheminformatics in R
- UE 4: Data analyses and data visualization in R

Due to the need for distant learning, the practical entity will be done in the form of either a Journal Club presentation or a small data science project by each student during the summer semester 2020.