Algorithms and Aspirins

While the human genome was sequenced over twenty years ago, we are still far from understanding it. How do different cell types in our body arise from the same DNA sequence? Why do some DNA mutations cause diseases while others do not? The real challenge does not lie in the sequence itself, but in how, where and when each part of the genome needs to be expressed. Today, the power of Artificial Intelligence can be used to crack this genetic code and is already pushing the boundaries of human knowledge. For instance, large language models similar to ChatGPT can learn the grammar of DNA sequences, and models like AlphaFold can accurately predict the structure of proteins. How are these technologies transforming biology and medicine? Can we interpret the biological principles behind these predictions? What are their limitations? I will discuss about these challenges and explain how we are using AI models to study the process of protein synthesis

Have you ever wondered how medicines end up in the apotheke? Drug discovery is a multidisciplinary endeavour at the intersection of biology, chemistry, and computer science. Today, AI models are used at almost every stage of this complex process. We will look at what this pipeline entails, and how AI is helping (also, where it fails), in the context of a deadly lung disease called idiopathic pulmonary fibrosis (IPF).