Proteins are the basic building blocks of all cells in living organisms. They are essential for the proper functioning of the whole organism, as they perform various functions - from muscle formation to the vital production of hormones and antibodies. They are particularly useful in many industrial processes, such as the production of food supplements or medicines.
“Disruption of protein structure due to harmful mutations can lead to loss of function and many serious diseases. To design unique drugs and vaccines, it is essential to decipher the relationship between their structure and function at the molecular level,” says Jiří Damborský from Loschmidt Laboratories of Protein Engineering at Masaryk University and St. Anne's University Hospital in Brno. Together with IT4Innovations National Supercomputing Center, which is part of VSB - Technical University of Ostrava, 14 Czech institutions are involved in the PerMed Centre. This research project aims to develop personalised diagnostic methods and drug candidates to help specific patient groups. The entire project is based on an interdisciplinary combination of medicine, chemistry, biology, and bioinformatics.
Modern research in natural sciences cannot do without supercomputers and specialised algorithms. The development of new bioinformatics tools and the enormous computational power of supercomputers make it possible to simulate thousands to millions of experiments that would be theoretically impossible to perform in laboratories. For example, determining the 3D conformation of a protein from its amino acid sequence requires months of experimental work and highly specialised technology. In contrast, computers can do it in minutes thanks to machine learning-based computational algorithms.
“Today, drug design can no longer be done without specialised tools and involves the use of supercomputers. All this saves significant costs and time. The use of supercomputers makes it possible to focus laboratory testing on the most promising compounds. Using these state-of-the-art technologies makes it possible to create the most effective drugs,” says Branislav Jansík from IT4Innovations National Supercomputing Center in Ostrava.
As part of one of the PerMed Centre's sub-projects, IT4Innovations National Supercomputing Center is developing a software package for protein modelling and analysis. The highly specialised tools focus on the computational study of proteins and significantly facilitate researchers' understanding of protein chemistry thanks to their speed and accuracy. "In practice, this means that the scientist puts the protein to be analysed into the software, which analyses it based on a large amount of data it contains. The protein is compared to other proteins using extensive databases, allowing the protein to be classified into one of two categories: aggregating and non-aggregating. Classification of protein aggregation is important for large-scale protein production, as well as analysis of the effect of mutations on the development of neurodegenerative diseases. Based on the calculations, tests will be carried out in the laboratories to verify the correct classification of the protein into a given category. Linking the software to the supercomputing infrastructure is important in terms of speed," say Jiří Damborský and Jan Martinovič about how the software works.
The supercomputing centre team collaborates with a team from Loschmidt's protein engineering labs on four software projects. Two are being developed in a completely new way, while the other two are improving existing solutions. The aim is to make the software easy and safe to use for all those involved in protein studies, without any prior knowledge of programming or computer science. "In addition to user-friendliness, developing solutions promise additional benefits. We strive to make these tools more informative, accurate, and reliable for researchers. We will achieve this by using the latest knowledge in machine learning and applying experience from previous projects we have been involved in," comments Jan Martinovič.
The software packages will be completed by the end of 2022 and then made available to researchers free of charge.
The PerMed Centre project is funded by the Programme for Support of Applied Research, Experimental Development and Innovation of the National Centre of Competence of the Technology Agency of the Czech Republic.