As the novel Coronavirus spreads, research is becoming digital in order to attain fast, and promising solutions. In fact, Google’s sister company, DeepMind is currently using Artificial Intelligence (AI) to study and understand the virus in order to speed up the efforts to develop a vaccine.

Learn the high demand technologies in a world-class learning platform for free

With investors including Elon Musk, Peter Thiel, Scott Banister, Jaan Tallinn, Horizons Ventures, and the Founders Fund — Demis Hassabis along with Mustafa Suleyman and Shane Legg founded DeepMind Technologies in 2010. Acquired by Google in 2015, DeepMind made it into AI’s history when it’s AlphaGo software was the first computer program to beat a professional human GO player, Fan Hui, in a series of games in October 2015.

DeepMind’s AlphaFold is essentially a protein structure prediction system. Its central component is the convolutional neural network that is trained based on the structures extracted from the Protein Data Bank. The PDB is a database for three-dimensional structural data of biological molecules like nucleic acids and proteins.

Proteins are large biomolecules that contain long chains of amino acids. To get to the bottom of a deadly virus, there is a demand for an intensive study to understand the structure of its protein. This is in order to interpret its function to create therapeutic drugs that work along with it. Traditional protein structure prediction methods essentially require techniques like X-ray crystallography, nuclear magnetic resonance, or cryo-electron microscopy. These methods are not only time-consuming but are also labour-intensive. However, by deploying the pattern-recognition capabilities of Deep Learning, scientists can significantly accelerate research to determine protein structures of COVID-19 and invent drugs for the same.

Also Read: 10 Ways How AI Helps Fighting the Coronavirus

DeepMind’s AlphaFold contains a trained neural network that predicts a protein structure – given its sequence, using stochastic gradient descent on the protein-specific potential. At the moment, the latest AlphaFold system uses a Machine Learning technique called free modelling to predict highly accurate protein structures even without the availability of similarly structured proteins. DeepMind has officially released the predicted structures under an open license in order to help SARS-CoV-2 researchers around the world to tackle the COVID-19 pandemic.

While its structure predictions have not been peer-reviewed or experimentally verified yet, the seriousness and sensitivity of the outbreak led to DeepMind to decide the release of the predicted structures in the hope of contributing to the development of its treatments. On the brighter side, its findings and predictions of COVID-19’s protein structure have cut down on months of human efforts and costs.

Are you curious to learn more about AI? Explore a career and upskill with Great Learning’s Artificial Intelligence and Machine Learning program!



Please enter your comment!
Please enter your name here

seventeen − 8 =