Google's AI Technology Solved 50-Year-Old Grand Challenge in Biology

Google's AI technology company, DeepMind has made a breakthrough by solving one of the greatest biology challenges: the 3D structure of proteins from its amino-acid sequence.

Called the AlphaFold, the DeepMind program outperformed approximately 100 teams in a biennial protein-structure prediction challenge called Critical Assessment of Structure Prediction (CASP). The results were announced on Monday, November 30, at the start of the conference that was held virtually.

John Moult, a computational biologist from Univerity of Maryland and founder of CASP in 1994 said that Alphafold's discovery is a big deal and in some sense, the 50-year-old problem is solved.

Game-Changer Discovery

According to Nature's report, the ability to accurately predict the 3D structures of protein from its amino-acid sequence is a big deal to biology and in the field of medicine. It would help scientists to understand the building blocks of cells and make drug discovery for various diseases easier.

AlphaFold first joined the CASP in 2018 and came on top against other participants. But this year, the London-based DeepMind was head-and-shoulders above other teams and performed incredibly well that scientists described it as something that could cause a revolution in biology.

Andrei Lupas, an evolutionary biologist at the Max Planck Institute for Developmental Biology in Tübingen, Germany, said that AlphaFold is a game-changer. It already helped him find the protein structure that his laboratory has been annoyed with for a decade and he expects that this DeepMind technology will alter how he works.

"This will change medicine. It will change research. It will change bioengineering. It will change everything," Lupas said.

AI and Protein Structures

TechCrunch reported that the test by AlphaFold showed that AI could accurately figure out the structure of proteins in just days, which is a very complex and crucial task to figure out how to fight diseases as well as solving bigger problems like how to break down ecologically dangerous or toxic waste.

The approach of DeepMind involved using an "Attention-based neural network system" that can continually refine its own predictive graph of possible protein structure folding based on its folding history to create accurate predictions.

Knowing how proteins fold is key to understanding how diseases are transmitted and how allergies work. The knowledge of folding proteins could help scientists alter it and halt the progress of infection or correct the mistakes in the folding that leads to neurodegenerative and cognitive disorders.

DeepMind's AI technology offers that by accurately predicting the folds in protein structure at a lesser time and resource-consuming process that will accelerate the change of pace in therapeutic progress in medicine making.

It could address major global threats, like the COVID-19 pandemic that the world is currently facing, by predicting viral protein structures to a high degree before the new pandemic begins because it will speed up the development of vaccines and treatments.