Data allows to explore and interpret processes. But rarely does one measurement give enough insight to fully understand complex processes like substrate binding and transport. Hence it is very important to analyse as much data from various methods as possible and learn from it. Machine learning algorithms look for patterns in big data sets which then allow us to make predictions about new data. The more diverse and bigger the data set is the more robust will be the outcome. Hence the core objective is to develop a cloud-based data pipeline that then allows us to run advanced machine learning algorithms to gain insight of thermodynamic and kinetics parameters driving substrate binding and transport in vesicles and cells.
The main objectives of project 8 are:
- Development of a data pipeline of protein-ligand binding measurements gained from different measurement methods
- Development of a software for automated data analysis
- Determination of affinity, enthalpy and entropy of ligand-protein interaction for ligands of GAT1, DAT, and LeuT