Computational characterization of protein-RNA interactions and implications for phase separation

Despite what was previously considered, the role of RNA is not only to carry the genetic information from DNA to proteins. Indeed, RNA has proven to be implicated in more complex cellular processes. Recent evidence suggests that transcripts have a regulatory role on gene expression and contribute to the spatial and temporal organization of the intracellular environment. They do so by interacting with RNA-binding proteins (RBPs) to form complex ribonucleoprotein (RNP) networks, however the key determinants that govern the formation of these complexes are still not well understood. In this work, I will describe algorithms that I developed to estimate the ability of RNAs to interact with proteins. Additionally, I will illustrate applications of computational methods to propose an alternative model for the function of Xist lncRNA and its protein network. Finally, I will show how computational predictions can be integrated with high throughput approaches to elucidate the relationship between the structure of the RNA and its ability to interact with proteins. I conclude by discussing open questions and future opportunities for computational analysis of cell’s regulatory network. Overall, the underlying goal of my work is to provide biologists with new insights into the functional association between RNAs and proteins as well as with sophisticated tools that will facilitate their investigation on the formation of RNP complexes A pesar de lo que se consideraba anteriormente, el papel del ARN no es solo transportar la informacion genetica del ADN a las proteinas. De hecho, el ARN ha demostrado estar implicado en muchos procesos celulares mas complejos. La evidencia reciente sugiere que los transcriptos tienen un papel regulador en la expresion genica y contribuyen a la organizacion espacial y temporal del entorno intracelular. Lo hacen interactuando con proteinas de union a ARN (RBP) para formar redes complejas de ribonucleoproteina (RNP), sin embargo, los determinantes clave que rigen la formacion de estos complejos aun no se conocen bien. En este trabajo, describire algoritmos que he desarrollado para estimar la capacidad de los ARN de interactuar con las proteinas. Ademas, ilustrare aplicaciones de metodos computacionales para proponer una maquinaria alternativa para el Xist lncRNA y su red de interacciones. Finalmente, mostrare como las predicciones computacionales pueden integrarse con enfoques de alto rendimiento para dilucidar la relacion entre la estructura del ARN y su capacidad para interactuar con las proteinas. Concluyo discutiendo preguntas abiertas y oportunidades futuras para el analisis computacional de la red reguladora de la celula. En general, el objetivo subyacente de mi trabajo es proporcionar a los biologos nuevas ideas sobre la asociacion funcional entre ARN y proteinas, asi como herramientas sofisticadas que facilitaran su investigacion sobre la formacion de complejos RNP.