COMPSTAT 2022: Start Registration
View Submission - COMPSTAT2022
Title: Bayesian approach for modelling rna transcription Authors:  Elena Sabbioni - Politecnico di Torino (Italy) [presenting]
Gianluca Mastrantonio - Politecnico of Turin (Italy)
Enrico Bibbona - Politecnico di Torino (Italy)
Guido Sanguinetti - Scuola Internazionale di Studi Superiori Avanzati - SISSA (Italy)
Abstract: Gene expression is regulated through the fundamental process of transcription, splicing and degradation, which can be modelled as an ODE system, whose parameters need to be estimated from experimental data collected by single-cell RNA sequencing. By this technique, biologists can take only a single snapshot of the cellular states: they obtain the counts of unspliced and spliced mRNA molecules, for each gene and for each gene altogether at the moment of the sequencing, which actually corresponds to different levels of maturity in the evolution of the different cells, and then the cells are destroyed. The aim is to reconsider part of the methods currently used to estimate the parameters of the model and to describe the evolution of some cells over different cell types, exploiting the level of expression of their genes and the concept of RNA-velocity. We reformulate it in a way that is mathematically better founded, using Bayesian statistics. We discuss the advantages of this approach in terms of the quality and the interpretability of the results.