UMR 7205 ISYEB Campus Jussieu Bâtiment A, 4eme et. pièce 429 Sorbonne Université
The team "Adaptation, Integration, Reticulation and Evolution" studies evolution, and develops new methods using evolutionary networks to represent, characterize and analyze evolutionary processes in prokaryotes, eukaryotes and moving elements. In particular we are interested in investigating the position of genes linked to aging and longevity in interaction networks, such as the human protein-protein interaction networks, to better understand the evolution of longevity.
This internship aims at developing and testing tools for this analysis.
The evolution origin of aging in metazoans is paradoxical: standard evolutionary theory strongly rejects the evolution of genetic programs for aging, as such programs would run against individuals’ fitness interest. Rather, aging is proposed to have emerged from a weakening of natural selection past the reproductive period. However, aging-regulating- and longevity-associated genes have been found in model species and are evolutionarily conserved. We aim to retrace the evolutionary history of these genes using information provided by interaction networks.
In interaction networks, nodes represent genes/mRNA/proteins and edges connecting nodes indicate physical interaction or co-expression. The topological position of particular genes in these networks can be characterized using different centrality metrics. Moreover, nodes can be labelled with phylogenetic (age of the gene or the protein domains) and functional (aging-regulating, longevity-associated, pro-aging...) information. Ultimately, a combined analysis of the centrality of these nodes and of their evolutionary origins, can be performed to propose evolutionary scenarios for the evolution of aging in metazoans.
Many tools have been developed in our lab to manipulate, analyze and label network graphs. The next step is to streamline network production and analysis with a dedicated pipeline.
- Develop a tool to standardize the assembly and labeling of networks from publicly available datasets, starting with human PPI or transcriptomes.
- Develop tools to extract network modules depending on user-defined combinations of labelling and centralities parameters.
- Perform preliminary comparative analysis between human and i.e. mouse networks and modules.
This work will be funded by the AIRE team.
M2 student, biostatistics, bioinformatic, math/info, ...
Python (networkx, iGraph modules), (R)
Strong interest for evolutionary biology