bioRxiv: Discover potentially nalization antibodies to new coronavirus using machine learning.

April 14, 2020 / / -- Viruses are cunning little pathogens that can wreak havoc on humans before our immune system knows how to destroy them.
machine learning tools, we can beat them by accelerating the formation of antibodies.
at Carnegie Mellon University's Department of Mechanical Engineering, Amir Barati Farimani developed algorithms that can infer, learn, and predict mechanical systems based on data.
he has studied a range of topics, from fluid mechanics and heat transmission to material discovery and robotics, as well as human health and bioengineering challenges.
Farimani is an assistant professor of mechanical engineering at Carnegie Mellon University, where he directs the Mechanical and Artificial Intelligence Laboratory.
: BioRxiv With the outbreak of the COVID-19 pandemic, Barati Farimani quickly shifted the focus of his lab to SARS-CoV-2 research.
previously used machine learning tools to study antibodies to Ebola and HIV, and now he wants to further study the new coronavirus.
scientists now use computational and physical models to screen thousands of antibody sequences.
these models are expensive and time-consuming, and we need information about SARS-CoV-2 that we don't already have.
"This is where machine learning can do the heavy lifting," says Barati Farimani.
not only 'learns' complex antigen-antibody interactions faster than current screening methods, but also exceeds the human immune system in response time," he said.
team integrated biological data from other existing infectious viruses into a data set they named VilusNet.
then used the data to train machine learning models and select the best-performing models to screen thousands of potential antibody candidates.
the model eventually identified eight stable antibodies, which are very effective in mesizing SARS-CoV-2.
the findings were published in a preliminary report on bioRxiv, a biology preprint server, so that other researchers could get the information as quickly as possible.
"Our goal is to save lives," Barati Farimani said.
"now share our preliminary findings that will help other scientists around the world in their efforts to combat the virus."
we share a common goal.
" () Reference: 1. Outsmarting a virus (2) Rishikesh Magar et al. Potential Neutralizing Antibodies Discovered for Novel Corona Virus Using Machine Learning, bioRxiv. (2020). DOI: 10.1101/2020.03.14.992156.