Amir Barati Farimani received his Ph.D. in 2015 in Mechanical Science and Engineering from the University of Illinois at Urbana-Champaign. His Ph.D. thesis was titled “Detecting and Sensing Biological Molecules using Nanopores.” He extensively used atomistic simulations to shed light on the DNA sensing and detection physics of biological and solid state nanopores. Right after that, he joined Professor Vijay Pande’s lab at Stanford. During his post-doc, he combined machine learning and molecular dynamics to elucidate the conformational changes of G-Protein Coupled Receptors (GPCRs). He specifically was focused on Mu-Opioid Receptors to elucidate their free energy landscape and their activation mechanism and pathway.
The Barati Farimani’s lab, the Mechanical and Artificial Intelligence laboratory (MAIL), at Carnegie Mellon University is broadly interested in the application of machine learning, data science, and molecular dynamics simulations to health and bio-engineering problems. The lab is inherently a multidisciplinary group bringing together researchers with different backgrounds and interests, including mechanical, computer science, bio-engineering, physics, material, and chemical engineering. The mission is to bring the state-of-the-art machine learning algorithm to mechanical engineering. Traditional mechanical engineering paradigms use only physics-based rules and principles to model the world, which does not include the intrinsic noise/stochastic nature of the system. To this end, the lab is developing the algorithms that can infer, learn, and predict the mechanical systems based on data. These data-driven models incorporate the physics into learning algorithms to build more accurate predictive models. They use multi-scale simulation (CFD, MD, DFT) to generate the data.
Outsmarting a Virus
Can machine learning help us to accelerate the antibody discovery process to fight highly infectious viral diseases and save thousands of lives?
Viruses are sneaky little pathogens that can wreak havoc on the human body before the immune system knows how to destroy them. Machine learning is a tool that can help us outsmart viruses by speeding up the process of developing antibodies.
My lab develops algorithms for machine learning that can infer, learn, and predict mechanical systems based on data. We can use these algorithms to learn the complex antigen-antibody interactions of viruses faster than the human immune system can. With Ebola, HIV, and SARS-CoV-2 (which causes COVID-19), this means potentially saving thousands of lives.
How does machine learning work for this? Well, scientists currently use expensive and time-consuming computational and physics-based models to screen thousands of antibody sequences. These methods require information that we might not yet have about a new virus. This is where machine learning can do the heavy lifting.
When COVID-19 emerged, our research team combined available biological data on other infectious viruses into a dataset. We used this to train machine learning models, selecting the best-performing model to screen thousands of potential antibody candidates. The model ultimately identified eight stable antibodies that were highly efficient in neutralizing SARS-CoV-2. We shared our findings with other scientists to help fight the pandemic.
The machine learning model we’ve developed is intended to assist biologists so they can zero-in quickly on the best antibodies to further investigate. Our model can also be particularly advantageous when new variants emerge, like the COVID-19 mutations in England and South Africa, or the recent Ebola outbreak in West Africa.
We’re working now to make predictions about the interplay of COVID-19 and factors such as the number of cases in each state based on the number of people who are vaccinated. Our ultimate goal is to save lives.