The way cell differentiate from bone marrow to peripheral blood level plays a crucial role in understanding and treating rare diseases and more common tumours. The main goal of this paper is to introduce a flexible statistical framework able to describe the cell differentiation process and to reconstruct a dependence structure along different levels of differentiation. We use next generation sequencing data on haematological diseases (severe combined immunodeficiency) within a gene therapy framework. The proposed statistical approach is based on Bayesian networks (BNs) and aims at finding a probabilistic model to describe the most important features of cell differentiation, without requiring specific detailed assumptions concerning the interactions among genes or the confounding effects of experimental conditions. Bayesian networks enable analyses on gene therapy‐treated patients in a data‐driven fashion and allow for exploring all relationships among different blood cell types integrating biological information, subject‐matter knowledge, and probabilistic principles.
Di Serio, C., Scala, S., Vicard, P. (2020). Bayesian networks for cell differentiation process assessment. STAT, 9(1) [10.1002/sta4.287].
Bayesian networks for cell differentiation process assessment
Paola Vicard
2020-01-01
Abstract
The way cell differentiate from bone marrow to peripheral blood level plays a crucial role in understanding and treating rare diseases and more common tumours. The main goal of this paper is to introduce a flexible statistical framework able to describe the cell differentiation process and to reconstruct a dependence structure along different levels of differentiation. We use next generation sequencing data on haematological diseases (severe combined immunodeficiency) within a gene therapy framework. The proposed statistical approach is based on Bayesian networks (BNs) and aims at finding a probabilistic model to describe the most important features of cell differentiation, without requiring specific detailed assumptions concerning the interactions among genes or the confounding effects of experimental conditions. Bayesian networks enable analyses on gene therapy‐treated patients in a data‐driven fashion and allow for exploring all relationships among different blood cell types integrating biological information, subject‐matter knowledge, and probabilistic principles.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.