How is protein decoding technology applied?

Research awarded the 2024 Nobel Prize in Chemistry could unlock some of the decades-old secrets of proteins, from discovering new drugs to enzymes that break down pollutants.

The 2024 Nobel Prize in Chemistry voting committee evaluated the protein research of three scientists as holding huge potential in a variety of fields, according to AFP. Biochemist David Baker, 62 years old, Professor at the University of Washington, USA, seeks to design a completely new type of protein never before seen in nature. While researchers Demis Hassabis, 48, and John Jumpe, 39, at Google’s DeepMind lab use artificial intelligence techniques to predict protein structure.

Proteins are molecules that act as “the factory of everything that happens in the body,” according to Davide Calebiro, a protein researcher at the University of Birmingham, UK. DNA provides the design for every cell. The protein then uses this information to transform the cell into a specialized type such as a brain cell or a muscle cell.

Protein consists of 20 different amino acids. The starting sequence of these acids determines the 3D structure into which they will twist and fold. American Chemical Society President Mary Carroll compares them to old landline telephone wires. “You can stretch a telephone wire and get a straight structure. Then it will spring back to its 3D shape,” Carroll said. So if chemists want to master proteins, they need to understand how 2D chains turn into 3D structures.

David Baker successfully used these building blocks to design a new protein unlike any other. Since then, his research team has continuously created innovative proteins, including proteins that can be used as pharmaceuticals, vaccines, nanomaterials and microscopic sensors.

The combination of AI

In proteins, amino acids are linked together into long chains that fold together to form a three-dimensional structure, which is decisive for the function of the protein. Since the 1970s, researchers have tried to predict protein structure from amino acid sequences, but this is extremely difficult. Many people failed to predict the structure of the new amino acid.

Up until 2020 Hassabis and Jumper trained their AlphaFold artificial intelligence model on every known amino acid sequence and corresponding structure. When encountering an unknown sequence, AlphaFold compares it with previous sequences, gradually reconstructing the structure using 3D images. The research team then developed an upgraded system, AlphaFold2, that helps predict the structures of all 200 million proteins that researchers have identified.

Since that breakthrough, AlphaFold2 has been used by more than two million people from 190 countries. Among a range of scientific applications, researchers could better understand antibiotic resistance and create enzyme images that can break down plastic.

Mastering proteins enables scientists to develop countless potential future applications. It allows people to better understand how life works, including why some diseases develop, how antibiotic resistance occurs or why some bacteria can break down plastic. Creating new proteins could lead to a variety of nanomaterials, targeted drugs and vaccines, or more environmentally friendly chemicals.

When asked about his favorite protein, Baker chose the one he designed during the pandemic, which can protect the body against the coronavirus. “I’m very excited about the idea of ​​a spray containing designer proteins, helping to block all epidemic viruses,” Baker shared.

They’re Khang (Theo AFP)


By Editor

Leave a Reply