Skip to ContentSkip to Navigation
Over onsNieuws en agendaNieuwsberichten

Extra Colloquium Mathematics, Professor Clelia Di Serio

25 juni 2014

Join us for coffee and tea at 10.45 a.m.


Wednesday, June 25th 2014


Prof. Clelia Di Serio,
Vita-Salute San Raffaele University


5161.0293 (Bernoulliborg)



Title: Exploiting NGS data from gene therapy treated patients to assess cell
        differentiation process: a graphical models approach.


The novelty of the present work is to use information arising from Next Generation Sequencing data in gene therapy frameworks as a point-wise “label” to target hematopoietic cell differentiation process. Indeed, one major challenge in complex systems biology is to provide a general theoretical framework that describes the phenomena involved in cell differentiation, i.e. the process whereby stem cells, which can develop into different types, become progressively more specialized. A number of biologically important processes involve transitions through distinct cell differentiation levels. Differentiation state changes in such processes  and are in general stochastic, as reflected in experimentally observed variation in transition latency even in the setting where transitions arise in homogenous cell cultures subjected to defined driving events. Many different biological models have been proposed in the literature. However no statistical evidence has been yet associated to biological proposals, and there is still a certain degree of uncertainty regarding the major branching directions characterizing hematopoietic differentiation as well as the main connection paths among cell types within the same differentiation level.
In this work we aim at providing  a statistical framework which is able to model hematopoiesis and to give evidence supporting specific hierarchical structures. A probabilistic modelling framework should 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,  typically induced by gene expression data.
We place this framework within a Bayesian Network context to be able to handle complex dependence structure among several variables combining information on conditional probability distributions and graphical models of dependencies.

Colloquium coordinators are Prof.dr. A.C.D. van Enter (e-mail : and
Dr. A.V. Kiselev (e-mail: )

Laatst gewijzigd:07 juli 2014 16:54

Meer nieuws