Network-based approach to design new strategies for cellular reprogramming: Clinical Applications

The goal of this project is to build a computational platform that relies on gene regulatory network models to develop strategies for identifying optimal reprogramming determinants triggering transitions between specific cell types. The platform is based on Boolean network modelling framework to simulate cell state transitions. Importantly, it is applicable for closely related cell sub-types with similar transcriptional profiles. This platform will be applied to projects relevant to cell therapies and regenerative medicine. For example, in collaboration with Professor Michele De Luca at the Centre for Regenerative Medicine in Modena, Italy, we intend to predict reprogramming determinants to derive corneal limbus stem cells from cultured epithelial keratinocytes. The derived experimental protocol will be further optimized and used for treating patients who have lost their corneal limbus stem cells by injury or burn.

Development of computational strategies for efficient cell conversions based on single cell RNA-sequencing

In our previous work, we developed a computational platform for identifying core transcription factors (TFs) that define the identity of each subpopulation of cells in a given single-cell RNA-seq dataset. The method predictions were validated experimentally for the conversion of human neuroepithelial cells into dopaminergic neuron progenitor cells. However, within the transcriptional core, identification of most efficient combinations of TFs (sub-cores) remains a problem. In the current project, we develop a computational method for identifying such sub-cores of TFs that can more efficiently induce cellular subpopulation conversions. Our approach will combine gene regulatory network inference with information theoretic measures of synergy and redundancy to identify sub-cores in the entire transcriptional core that determines subpopulation identity. This project is carried out in collaboration with Prof. Deepak Srivastava’s lab at the Gladstone Institute, where our predictions for efficient conversions of cardiac right ventricular cells into left counterparts will be validated. The outcome of this project will be essential for designing strategies to treat patients with cardiovascular diseases.

Computational approach to directing cellular reprogramming into multiple cell types

Current reprogramming protocols are able to reprogram one cell type into another. However, derivation of in vitro tissues consisting of multiple cell types from a single source of cells is a challenge and is of clinical interest. In this project we aim to develop computational methods that predict reprogramming factors that can convert one cell type into multiple cell types in a controlled way. Our previous work for identifying core TFs in each subpopulation of cells from single-cell RNA-seq data will be extended, so that it can predict an optimal set of TFs that could simultaneously induce multiple reprogramming from one cell type. The experimental validation will be performed in collaboration with Prof. Ernest Arenas’ lab at Karolinska Institute. Experiments on cultured astrocyte cell colonies will be conducted to perturb sets of TFs that have been predicted to induce cellular conversions into different proportions of neural stem cells and neurons. The outcome of this project will be used for transplantation therapies for neurodegenerative diseases, such as Parkinson’s disease. This project is supported by the Luxembourg National Research Fund (FNR).

Integrative gene regulatory network reconstruction for reverting the pathologic phenotype of Alzheimer’s Disease

Alzheimer’s disease (AD) is one of the most prevalent neurodegenerative diseases worldwide, whose development and onset has been linked to epigenetic and transcriptional dysregulations in the brain. However, it is not yet understood how these alterations could be reverted to restore a healthy cellular program. In this project, we aim to identify the minimal number of genes required to be perturbed in order to revert the pathologic phenotype. Our approach integrates transcriptomics and DNA methylation data from the middle temporal gyrus in the brainstem into a gene regulatory network model of age-matched healthy and pathologic individuals for simulating the cellular response to perturbations. Experimental validation of the predictions will be performed in a novel AD model, which is based on induced pluripotent stem cells from pathologic donors. This project is carried out within the EPI-AD consortium and is involving partners at Maastricht University (Netherlands), the University of Exeter (UK), the University of Würzburg (Germany), and Bellvitge Biomedical Research Institute (Spain).