Scientists rely on artificial intelligence modeling to understand more about protein-sugar structures

Scientists rely on artificial intelligence modeling to understand more about protein-sugar structures

The sugars linked with the reported software are a very good match with both AlphaFold and experimental protein models. Credit: Dr Jon Agirre

New research based on artificial intelligence algorithms has enabled scientists to create more complete models of the protein structures in our bodies, paving the way for faster design of therapies and vaccines.

The study, conducted by the University of York, used artificial intelligence (AI) to help researchers better understand the sugar that surrounds most of the proteins in our body.

Up to 70% of human proteins are surrounded or stacked with sugar, which plays an important role in how they look and act. Furthermore, some viruses such as those underlying AIDS, influenza, Ebola and COVID-1

9 are also protected by sugars (glycans). Adding these sugars is known as a modification.

To study proteins, the researchers created software that adds missing sugar components to models created with AlphaFold, an artificial intelligence program developed by Google’s DeepMind that makes predictions on protein structures.

Senior author, Dr Jon Agirre of the Department of Chemistry, said: “Human body proteins are tiny machines that make up our flesh and bones by the billions, carry our oxygen, allow us to function and they defend us from pathogens. And just as a hammer relies on a metal head to hit sharp objects, including nails, proteins have specialized shapes and compositions to do their job. “

“The AlphaFold method for protein structure prediction has the potential to revolutionize workflows in biology, enabling scientists to understand a protein and the impact of mutations faster than ever.”

“However, the algorithm does not take into account the essential changes that affect the structure and function of proteins, which gives us only part of the picture. Our research has shown that this can be addressed relatively simply, leading to a more complete structural forecast. “

The recent introduction of AlphaFold and its protein structure database allowed scientists to have accurate structure predictions for all known human proteins.

Dr. Agirre added: “It’s always good to see an international collaboration grow and bear fruit, but this is just the beginning for us. Our software has been used in the structural work of glycan that has supported mRNA vaccines against. SARS-CoV-2, but now there is much more we can do thanks to AlphaFold’s technological leap. We are still in its infancy, but the goal is to move from reacting to changes in a glycan shield to anticipating them. “

The research was conducted with Dr Elisa Fadda and Carl A. Fogarty of Maynooth University. Haroldas Bagdonas, Ph.D. student at the York Structural Biology Laboratory, which is part of the Chemistry Department, also worked on the study with Dr. Agirre.

The document is published in Structural and molecular biology of nature.


DeepMind and EMBL release the most comprehensive database of predicted 3D structures of human proteins


More information:
Bagdonas, H. et al, The case of post-predictive modifications in the AlphaFold protein structure database, Nat Struct Mol Biol (2021). doi.org/10.1038/s41594-021-00680-9

Provided by the University of York

Citation: Scientists rely on AI modeling to understand more about protein-sugar structures (2021, November 2) recovered November 2, 2021 from phys.org/news/2021-11-scientists-ai-protein -sugar.html

This document is subject to copyright. Except for any correctness for private study or research purposes, no part may be reproduced without written permission. The content is provided for informational purposes only.


Read more here: Source link