Artificial Intelligence from Google studied the structure of coronavirus

US, WASHINGTON (NEWS OBSERVATORY) — DeepMind, Google’s artificial intelligence (AI) division, has joined the global research community studying the new coronavirus COVID-19.

DeepMind is best known for its AI, which easily defeated the world’s best players in Go and StarCraft II. The research lab is currently using its system to help researchers deal with the epidemic.

To study the virus and develop a vaccine, scientists must first understand how it functions, namely the structure of the viral proteins. This is a lengthy process that takes months and may not always produce results. Scientists turned to computer predictions using a deep learning system known as Alpha Fold.

Coronavirus work is being done in laboratories around the world. DeepMind hopes to help these studies “by releasing structural predictions of several little-studied proteins associated with SARS-CoV-2, the virus that causes COVID-19.” The system uses a machine learning method without modeling the environment, with which one can predict protein structures in the absence of similar protein structures.

DeepMind employees hope to save scientists the months that are usually spent on determining the protein structure of the virus. “Knowing the structure of a protein provides an important resource for understanding how it functions, but experiments to determine the structure can take months or more,” the company’s official blog says.

Given the “potential seriousness and time frame”, DeepMind said it was going to skip the process of experimentally validating the results or waiting for the academic community to review them before publishing. This is consistent with other scientific studies on the topic, which appear both in peer-reviewed journals and in preprints without peer-review, as this process can take months.

“We emphasize that these predictions of the structure have not been experimentally verified, but we hope that they can contribute to the work of the scientific community on how the virus works and serve as a platform for generating hypotheses for future experimental work on the development of therapeutic agents,” said in a blog post.

The team notes that the data provided “are not the main focus of current therapeutic activity”, but can help a common understanding. “It is important to note that our structure forecasting system is still under development, and we cannot be sure of the accuracy of the structures that we provide, although we are confident that the system is more accurate than our previous CASP13 system. We have confirmed that our system provides an accurate prediction for the experimentally defined SARS-CoV-2 structure stored in the Protein Data Bank, and it gives us confidence that our model predictions for other proteins can be useful, “the researchers say.

An open license will allow any researcher to develop, adapt or share the results of DeepMind research. Google acquired London’s research organization DeepMind for £ 400 million back in 2014. The company had previously used AI for health research, developing models for identifying eye diseases and detecting neck cancer.

Alibaba is also researching coronavirus. So, researchers from a Chinese corporation announced the development of a machine learning algorithm that allows to detect, with an accuracy of 96%, pneumonia caused by the new coronavirus COVID-19, distinguishing it from inflammations of a different nature. According to the Nikkei Asian Review, analysis will require computed tomography of the patient’s chest. After analyzing the image for 20 seconds, the system gives an answer – for this the doctor would need numerous images and at least 15 minutes of time.

The algorithm was trained on 5 thousand images of lungs of patients with confirmed coronavirus infection and is already used in at least 100 hospitals throughout China.


Our Standards, Terms of Use: Standard Terms And Conditions.

Contact us: [email protected]

Article is written and prepared by our foreign editors from different countries around the world – material edited and published by News Observatory staff in our US newsroom.