Recently, diseases related to obesity have emerged as a serious and increasing public health issue worldwide. Indeed, adipose tissue is central in many physiological pathways involved in these diseases. Clarifying the dynamics of adipocyte size and number evolution is crucial to better understand the pathophysiological basis of those related diseases. We will study adipocyte size dynamics with mathematical models based on biological data, in particular in various health conditions (healthy, obesity, kidney diseases). We aim to mimic characteristic cell size distributions and therefore provide information on their connections with health conditions. This highly interdisciplinary project builds on a strong collaboration between biologists, physicians and applied mathematicians. The adipocytes are designed to store energy in form of lipids by adapting their size to accommodate the storage. When intake is much larger than release, adipocytes compensate with two mechanisms. First, the cell sizes grow; second, the number of adipocytes increases. In terms of dynamics, no exact connection between increase of the number of adipocytes and growth of their size is known. In addition, when fat mass increases drastically, the adipose tissue experiences remodeling, especially the extracellular matrix with fibrosis development. Bariatric surgery has become the major treatment against severe obesity. It aims at limiting the amount of food intake by reducing gastric and intestinal tracts. After this surgery, weight loss is correlated with a reduction of adipocyte average size. Our goal is to develop a mathematical model that describes the adipocyte size dynamics, based on exchange of lipids, including the recruitment of new adypocytes and the constraints of fibrosis development. The available data will be used to parameterize the model through adapted parameter estimation procedures that we will develop. We will describe the time evolution of adipocyte size, in control situation for several animal species, after fat diet or caloric restriction, and in obese patients that underwent bariatric surgery. First, we will compare model parameter values in different species to identify the model parameters that are the most species-dependent. Then, we aim at highlighting in our model the main mechanisms that are impacted by health conditions. Moreover, based on numerical simulations we expect to predict patient follow-up after bariatric surgery, and explore numerically new treatment strategies. Finally, we expect to design new experiments to validate our modeling hypotheses. Altogether we expect our research results to have a significant impact on the society with an improved health care for obese patients. Using numerical and mathematical tools to assist clinical decisions represents an advantage regarding patient health as well as health care cost reduction and it presents collaboration opportunities with industrial partners to manufacture these tools.

Evolutionary innovations can occur via several mutational paths but an integrated view of novel gene acquisition at the population level is still lacking. The different mechanisms that generate or introduce new genes in a genome act continuously throughout evolution, resulting in genes of different ages. However, most studies on novel genes were conducted at the interspecific level by comparing a single reference genome per species. Population genomics is presently shifting the field of comparative genomics from single reference genomes to population pangenomes, thereby giving access to individual variations in presence/absence of genes at the population level. The collection of genes present in a population is named the pangenome. The pangenome consists of core genes invariably present in all individuals and accessory genes that are segregating in the population at varying frequencies. Here, we define Novel Accessory Genes (NAGs) as the subset of accessory genes that are not vertically inherited from the species ancestor but were gained or emerged during the diversification of the species. NAGs originate mainly from introgression events, horizontal gene transfers (HGTs) and de novo gene emergence. When novel genes first appear, they are present at very low frequency in a population, likely in a single individual, and subsequently can either disappear or eventually raise in frequency. Therefore, we make the hypothesis that the best evolutionary time scale to investigate gene acquisition mechanisms would be at the population level. We propose to use the yeast Saccharomyces cerevisiae as a model organism to explore, at the species level and at the genome scale, the dynamics of gene acquisition at the time they arise and before they are removed by selection. We will directly benefit both from the high quality population genomic dataset available and the possibility to perform large-scale experimental testing to a level not accessible in any other model organism. The main goal of our proposal is to systematically explore the mechanisms of acquisition of NAGs and their relative contributions to the emergence of evolutionary novelties. First, we will analyse the evolutionary trajectories of NAGs based on high quality population genomics datasets and using state of the art computational approaches. Second, we will measure fitness effect of natural and engineered NAGs by combining genome editing, synthetic biology and high throughput phenotyping. Finally, we will investigate how NAGs functionally wire into cellular networks by measuring transcription, translation and post- translational modifications for both natural and engineered NAGs in various environments in order to quantify their level of i cellular integration. Altogether we plan to generate a large pool of heterogeneous data that we will integrate into a single conceptual framework including information on ecological niches and genetic backgrounds. The outcome of this proposal will provide a multi-layered view of the functional and evolutionary impacts of NAGs in the yeast pangenome revealing general rules leading to the evolution of new genes and new functions in eukaryotes.

This project aims at developing MinOmics, an integrated analysis and visualization framework dedicated to multi-omics data analyses and focused, as a study case, on redox signaling networks. We propose an immersive visual analytics approach where experimental data and analysis results can be visualized and manipulated interactively on an 8.3 m2 large-scale 25 MPixel high-resolution display wall. Modeling of the redox network will be based on original mathematical solutions to reduce the number of parameters to estimate. Structural interpretations will be enabled by an automated modeling pipeline for the prediction of modified sites. Bioinformatics modeling will allow us to uncover the structural determinants of the different modifications and to unravel the dynamics of the redox network and may open new areas of research relevant for agriculture, biotechnology and medicine. This project addresses how to take into account state-of-art modeling in the context of integrative biology.

The repair of DNA lesions involves at early stages an active remodeling of the chromatin architecture. If the exact function of this mechanism remains unclear, it is essential for an efficient correction of the DNA breaks. The poly-ADP-ribosylation signaling pathway, and in particular the PAR-dependent remodelers, play a central role during these chromatin reorganization mechanisms associated with the activation of the DNA damage response of the cell. During this project, we aim not only at establishing a full description of these chromatin remodeling processes but also at better understanding their impact on the subsequent DNA repair steps. In this context, we will focus more specifically on the role of the PAR-dependent remodelers. To reach this goal, we setup a consortium gathering three research teams with complementary expertise. This consortium will establish a multi-scale analysis framework allowing to characterize the different folding levels of the chromatin: from the nucleosomes to the topologically associated domains (TADs). Dynamics at the nucleosome scale will be assessed by combining ATAC-seq with single-molecule imaging techniques: fluorescence correlation spectroscopy and single molecule tracking. At higher scales, the dynamics of chromatin loops and TADs will be monitored by chromatin conformation capture (4C-seq) and by tracking the movements of single TADs in the nucleus of living cells. These different methods will be used to analyze the changes affecting the chromatin structure in the vicinity of DNA breaks induced by laser irradiation or by restriction enzymes. Next, we will analyze the role of the PAR-dependent remodelers during these modifications of the chromatin structure. Finally, we will study how chromatin remodeling events specific to these remodelers influence the subsequent steps of the repair mechanisms. We will analyze in particular their impact on the regulation of the DNA end resection at the double-strand breaks. The switch between the initiation of this process or its inhibition determines whether the break will then be repaired by homologous recombination (HR) or by non-homologous end-joining (NHEJ). We will analyze the impact of knocking-down the PAR-dependent remodelers on the recruitment of the proteins regulating the initiation of the resection process. We will also monitor how these knock-downs affect the HR/NHEJ balance and influence the efficiency and the fidelity of the repair process. Altogether, these different results will allow us to uncover the specific function of the early chromatin remodeling mechanisms in the maintenance of genome integrity. By contributing to a better understanding of the role of the PARylation signaling pathway during the DNA repair process, our project may contribute to identify new therapeutic targets for cancer treatments. Moreover, given that many tumor cells display a perturbed chromatin architecture, our project will help to establish the impact of these perturbations on the efficiency of the repair mechanisms in these cells.

Monogenic diseases are due to mutations that affect 7,000 different genes. One major roadblock in the diagnosis of rare diseases is the number of gene variants of uncertain significance. Novel methodologies need to be developed specifically to determine the pathogenicity of variants in the coding sequence. This project focuses on the MEFV gene mutated in familial Mediterranean fever. Thanks to synthetic biology and DNA bar-coding, in vitro assays combined to deep sequencing, large scale in silico analyses and validation using primary cells from patients, we will determine the comprehensive repertoire of pathogenic MEFV variants. Finally, we will develop a web-based platform to provide the information on each MEFV variants to clinicians and geneticists. This project should provide a proof of concept that this innovative approach, can identify simultaneously all pathogenic variants in the coding sequence of a given gene with implications for numerous monogenic diseases.