Scientists develop AI modeling to better understand about protein-sugar structures

Scientists develop AI modeling to better understand about protein-sugar structures

The sugars attached to the reported software are a very good match to both AlphaFold and experimental protein models. Credits: Dr. Jon Agirre

New research building on AI algorithms has allowed scientists to create more complete models of protein structures in our bodies — paving the way for faster design of therapeutics and vaccines.

The study — led by the University of York — used artificial intelligence (AI) to help researchers better understand about the sugar that surrounds most of the proteins in our bodies.

Up to 70 percent of human proteins are surrounded or scaffolded by sugar, which plays an important part in how they look and behave. Moreover, some viruses such as those behind AIDS, Flu, Ebola and COVID-1

9 are also protected behind sugars (glycans). The addition of these sugars is known as modification.

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

The senior author, Dr. Jon Agirre from the Department of Chemistry: “The proteins of the human body are tiny machines that in their billions, make up our flesh and bones, carry our oxygen, allow us to function, and protect us from pathogens… And like a hammer that relies on a metal head to strike sharp objects including nails, proteins have special shapes and compositions to do their jobs. “

“The AlphaFold method for predicting protein structure has the potential to change workflows in biology, allowing scientists to understand a protein and the impact of mutations more quickly than ever before.”

“However, the algorithm does not take into account the important changes that affect the structure and function of the protein, giving us only part of the picture. Our research has shown that this can be addressed in a fairly straightforward way, leading to a more complete structural prediction. ”

The recent introduction of AlphaFold and the accompanying database of protein structures has allowed scientists to make accurate structural predictions for all known human proteins.

Dr. added. Agirre: “It’s always nice to watch an international collaboration grow to bear fruit, but this is just the beginning for us. Our software has been used in glycan structural work that has adopted mRNA vaccines against SARS-CoV-2, but now we can do a lot more thanks to the AlphaFold technological leap. These are still early stages, but the goal is to move from responding to changes in a glycan shield to relying on them. “

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

The paper was published in Nature Structural and Molecular Biology.


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


Additional information:
Bagdonas, H. et al, The case for post-predictional changes 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

Sipi: Scientists develop AI modeling to better understand about protein-sugar structures (2021, November 2) retrieved November 2, 2021 from phys.org/news/2021-11-scientists -ai-protein-sugar.html

This document is subject to copyright. Other than any fair dealing for the purpose of private study or research, no part may be reproduced without written permission. Content is provided for informational purposes only.


Read more here: Source link