Abstract: In the central dogma of molecular biology, RNA is the messenger that deciphers DNA (the recipe) in order to create proteins (the cake). This "man in the middle" has a crucial function to maintain order in the system, it is the first molecule on which cells can play to adapt to sudden changes like stress, drug treatment, bacterial infection, etc. Understanding the stability of this molecule in the system is thus crucial to understand the cells well-being, and has direct clinical implications. In this talk I will first present a statistical model to estimate degradation rates from long-read RNA-sequencing data, and show how we can compare stability in different conditions. This is joint work with Nikolay Shirokikh at UWA. In a second part, I will present some very on-going project to model the processes leading to degradation, that are also studied using long-read RNA sequencing data. Here I will describe our first slightly naive model (this is joint work with Hoel Queffelec, Nikolay Shirokikh, and Dominique Gagliardi and Hélène Zuber at CNRS in Strasbourg), and hope to foster some novel collaborations with some stochastic PDE specialists to propose a more realistic model that also allows to analyze what happens outside of the steady state.