4 place Jussieu
The development of Acute Myeloid Leukemia (AML) is a progressive invasion of the bone marrow (BM) that parallels a hierarchical process of mutation acquisition. During this process, the crosstalk between normal or mutant Hematopoietic Stem and Progenitor Cells (HSPCs) and the BM environment may undergo multiple changes. The first lesions occuring in the AML clone (TET2/DNMT3A) alter their surrounding environment, including mesenchymal stromal cells (MSCs). In turn, modified MSCs may favor the expansion of AML cells. Understanding and modelling the crosstalks between these cell populations at time of disease initiation or evolution will help identifying new therapeutic tools against AML. This work is part of a project whose aims are to decipher and model the mechanisms of the crosstalk between normal, pre-leukemic, leukemic HSPCs and MSCs to further identify new targets to eradicate the malignant clone in AML. The experimental setup is divided in two parts: first a proof of concept, and then a study on patient samples. The proof of concept models the crosstalk between HSPCs engineered to mimic pre-leukemic lesions (TET2/DNMT3A mutations) and surrogate MSC cell lines. The patients samples are composed of primary HSPCs and MSCs from AML and healthy BM samples.
The goal of the postdoctoral project is first to reassess and perform a more in depth analysis of the set of RNA-Seq data that have been generated on the proof of concept and the patient samples (total: 80 samples). The preliminary analysis have highlighted groups of regulated genes that updated our view on the onset of AML and identified multiple interesting candidates related to epigenetic and non coding RNA regulation. The candidate will first strengthen those results and detail our understanding of the mechanism by analyzing more in detail the properties. If possible a first analysis about the effects of differential splicing will be made as well (Salmon).
Then in a second step, single cell RNA-Seq data will be generated, together with methylation data on the samples. The postdoctoral candidate will use and develop new methods that can (1) infer temporality from the single cell RNA-Seq (as donne by Monocle2 or Velocyto), (2) leverage the information from bulk RNA-Seq analysis and finally (3) combine it with the methylation data do understand the regulatory effect of methylation.
Finally the candidate will propose a method that can regress a vector of gene expressions into one or two real valued variables. Those variables are used within the team of collaborators (D. Sallort's team at LCQB) as descriptors of dynamical systems that provide a phenomenological insight on the landscape of tumoral progression scenarios.
The candidate should have an first experience with the analysis and manipulation of high throughput biological data (ideally genomics and transcriptomics).
Fluent in programming languages such as R or Python. Good knowledge of multidimensional data analysis technique.
Experience with single cell data would be a plus.
The postdoctoral candidate would work with two laboratories at the IBPS (institute Biology Paris Seine): Laboratory of Developmental Biology (head: T. Jaffredo) and the Laboratory of Computational and Quantitative Biology (head: A. Carbone).