As the world faces COVID-19, every bit of technological innovation and ability deployed to fight the pandemic has brought us one step closer to combat it. Artificial intelligence and machine learning are playing a key role in understanding the COVID-19 crisis. Multitudes of research teams around the world are integrating their efforts to collect data and develop solutions. Machine learning is used to diagnose patients to identify those at the highest risk of being infected. It applies in researching existing drugs that can help in reducing its effect, increase the pace of development of drugs for its treatment. It forecasts the spread of disease, map the path of viruses, and understand the coronavirus better.
Mary Jo Ondrechen and Penny Beuning, professors of Chemistry and Chemical Biology at the Northeastern University are using machine learning algorithms to identify all of the amino acids responsible for the abilities of coronavirus to infect and thrive at the expense of human cells. Proteins are large, complex molecules that do most of the work in the cells. They are made up of thousands of smaller units called amino acids and function through cascading interactions. These interactions often take place outside the place of chemical reaction, but can still control the action of different proteins and may cause hindrance in any specific chemical reaction. The algorithm developed by Ondrechen predicts many of these interactions based on the specific molecular structure of proteins. It can be used to gain a better understanding of the biochemistry of coronavirus.
The proteins in the coronavirus facilitate its ability to infect cells of the human body without resulting in any visible symptoms for a remarkable period. Many researchers are investigating the roles of each protein contained in coronavirus at the molecular level to inhibit those chemical reactions, which enable it to bind itself with other proteins on the surface of human cells and replicate.