Een compleet model voor eiwit aggregatie en zijn toepassing op het ontwerp van monoclonale antilichamen en peptides voor doelgerichte aggregatie

Protein aggregation, the process by which polypeptide molecules form larger order homotypic interactions, plays a major role in both healthy and and disease-associated physiology. Functional amyloids for example, have been discovered in all kingdoms of life, including human. In sharp contrast, depositions of protein aggregates in patient’s tissues are the clinical hallmark of over 30 human diseases and unwanted protein aggregation is an important risk in the application of proteins for therapeutic and other applications. Thanks to intense research efforts in the last decades and advances in experimental techniques, the field has made remarkable progress in recent years in understanding the molecular mechanisms that underlie protein aggregation. The field of computer science on the other, witnessed major advancements in the modelling of complex systems, and in particular machine learning ‘artificial intelligence’ has received a lot of attention. With the current proposal, I want to capitalise on the advances in both fields and create a general mathematical model to predict aggregation in proteins and peptides with an accuracy that surpasses anything that has been done thus far and that is capable of taking into consideration the environmental conditions in which they are.

Trefwoorden:Protein aggregation, Machine learning, Neural Networks, Peptide design, in silico mutagenesis

Disciplines:Proteïnen