I spent close to 20 years focused on mathematics and data science, including cofounding the research lab fast.ai, which focused on the powerful family of AI algorithms known as deep learning. A few years ago, I decided to make a big pivot and return to school for immunology. What motivated my sudden change amidst a successful career? We are now living in a pandemicene, a period with increasingly likely pandemics. Climate change and habitat destruction are crowding species into closer and closer contact with humans. Frequent global travel and mega-cities allow unprecedented opportunities for viruses to spread and mutate. Antibiotic resistance is rising rapidly.
At the same time, we have been learning more and more about the long-term consequences of infections– seemingly mild infections can contribute to long-term autoimmune diseases, neurodegenerative diseases, and cancer. The human immune system is incredibly complex and there is much we don’t know about it. Immunology is a crucial area to study. In this post, I want to gather some of my writing and talks on how AI is being applied to immunology.
My 30 minute presentation on key areas where deep learning is being applied to immunology
Using AI to predict what T cells will bind to
T cells are one of the most important cell types of our immune systems. Figuring out how to predict what a T cell will bind to (meaning what cells it can recognize as bad and coordinate attacks against) would be a vital medical breakthrough.
Mapping immune cell communication networks
Immune cells talk to each other through complex networks. NLP, math, and immunology are useful in studying these networks.
Discovering new antibiotics
Bacterial resistance to existing antibiotics is an urgent threat. AI is being used to search for new antibiotics.
Ethical risks of AI applied to immunology
The enthusiasm about AI in medicine is failing to grapple with realities of the system. Recognizing gaps in how AI is applied to scientific research can prevent potential shortcomings and risks.
You can subscribe to be notified of new blog posts by submitting your email below:
I look forward to reading your responses. Create a free GitHub account to comment below.