Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture delves into predicting protein structure from sequence data using Direct Coupling Analysis (DCA) and inferring interaction partners through the Iterative Pairing Algorithm (IPA). The DCA method allows for the full prediction of protein 3D structure, with a focus on evolutionary inferred contacts and folding calculations. The limitations and performance of DCA are discussed, highlighting its effectiveness in predicting 3D contacts. The lecture also covers co-evolution and correlations between interacting partners, using bacterial two-component systems as a dataset. The DCA-based method, IPA, is explained in detail, emphasizing correlations, direct couplings, and interaction energies. The lecture concludes with insights on the impact of sequence covariation on protein-protein interactions and future perspectives on improving PPI prediction and understanding the role of historical contingency in protein sequences.