We can define viruses as extremely tiny criminals which aim to destroy their host’s body, and sometimes the viruses accomplish their mission before our immune system can find the way to eliminate them and saves our lives. Hence, Carnegie Mellon University (CMU) is working on the project aiming at speeding up the process of developing antibodies with the help of machine learning.
As there are none of countries safe from the outbreak of the COVID-19 pandemic, Amir Barati Farimani from CMU’s Department of Mechanical Engineering wants to apply his expertise on algorithms to support doctors and fight against the virus.
With machine learning tools that were previously used for studying antibodies for viruses like Ebola and HIV, Barati Farimani hoped to gain better understanding of the novel coronavirus. Similarly, machine learning may be more effective than the computational and physics-based models which scientists are using currently because these models require information about SARS-CoV-2 that we do not have now.
“This is where machine learning can do the heavy lifting,” said Barati Farimani. “Not only can it ‘learn’ the complex antigen-antibody interactions much more quickly than the current screening methods, it can also beat the human immune system in response time.”
With VirusNet—the dataset that came from the combination of available biological data and other infectious viruses, the researchers could train machine learning models, and selected the one that delivers best performance to screen many potential antibody candidates.
According to the team, “the model could eventually identify eight stable antibodies which were efficient in neutralizing SAR-CoV-2.” The findings were posted in a preliminary report on the open-source outlet bioRxiv so that other researchers would have access to the information as soon as possible.
“Our goal is to save lives,” said Barati Farimani. “Sharing our preliminary findings now will help others scientists around the world in their work to fight this virus. We have the same goal.”
Retrieved from: https://engineering.cmu.edu/news-events/news/2020/04/10-outsmarting-a-virus.html
.jpeg)
