research 56 research 57 PROJECTS New research FNRS FRANCK DEQUIEDT Cancer-specific subversion of mRNA degradation pathways by a Prion-like oncogenic transcription factor unravels a new vulnerability in Ewing sarcoma Ewing sarcomas (EwS) are highly aggressive pediatric cancers with only 50 to 70% of survival 5 years after diagnostic, mainly due to metastasis and lack of targeted therapy. EwS are typified by chromosomal translocations that produce oncogenic fusion proteins, which are often the only recurrent genetic abnormality. These fusions are the key drivers of disease pathogenesis and are essential for the growth and survival of EwS cells. While the causality of these fusions in EwS is well established, their functions remain poorly understood. To date, EwS fusions have been mostly considered and studied as aberrant transcription factors. In this project, we will assess the provocative hypothesis that in addition to their role in transcription, EwS fusions subvert mRNA degradation pathways to impose a specific transcriptional landscape in EwS cells, that contributes to their role in cancer initiation and progression. We will also test the idea that this new function of EwS fusions provides novel therapeutic opportunities for this rare and still formidable family of malignancies. TOM DRUET ROAGE : Revealing past inbreeding dynamics in animal populations through improved estimation of age of homozygous-by-descent segments and runs-of-homozygosity (ROH) Homozygous-by-descent (HBD) segments occur when a DNA segment is inherited twice from a common ancestor through different genealogical paths, as a result of inbreeding. These segments appear as long stretches of homozygous genotypes known as runs-of-homozygosity (ROH). ROH are widely used for several applications in animal genomics, including estimating inbreeding coefficients, studying inbreeding depression, assessing genetic diversity, inferring past demography, and identifying selection signatures. The length of ROH is related to the number of generations to the common ancestor (i.e. it indicates the age of the inbreeding event) and can be used to estimate past effective population size (Ne) or to study the relationship between inbreeding age and mutation load. Despite the increasing popularity of these applications, there is still a lack of evaluation of their accuracy. We have recently shown that ZooRoH, a model-based approach we developed, is more efficient in assigning HBD segments to the correct age class. Comparisons with state-of-the-art methods are also lacking for the estimation of past Ne. Here, we will first use simulations to evaluate the accuracy of ZooRoH and ROHbased approaches for estimating the age of HBD segments. We will then implement new methodological features to improve ZooRoH, and develop a new approximate Bayesian computation framework to more accurately estimate past Ne with ROH-based approaches. These new methods will be evaluated on simulated data and on real data from populations with well recorded demographic history, including temporal genomic data. Finally, we will define new parameters derived from the distribution of HBD segments to characterize inbreeding dynamics, and then run ZooRoH on a large number of populations. The inferred parameters will be used to classify the populations into major groups related to their demographic history. This classification approach could then be applied to new populations of unknown status.
RkJQdWJsaXNoZXIy MTk1ODY=