Micro-organisms used for bio-based chemistry are studied with the aim of conceiving potential microbial factories. They are piloted by their genome expression, with very diverse mechanisms acting at various biological scales, sensitive to external conditions (environment, temperature, nutrients). The eruption of novel high-throughput experimental technologies has have demultiplied the available omicsdata and means of understanding for the studied systems. Their handling however increasingly require advanced bioinformatic tools based on optimization, graphs and machine learning strategies for instance, as well as theoretical frameworks to provide insights into the multi-level causation (Editorial: Multi-omic data integration). The global objective of the whole project is to develop innovative analysis methods for such highly integrated data modeled as networks. They should include genomic, transcriptomic and epigenetic data, for multiple microorganisms strains. The methodology would inherit from a wealth of techniques developed over graphs for scattered data, social networks, etc. and build upon biologically-related a priori as recently developed at IFP Energies nouvelles (BRANE Cut: biologically-related a priori network enhancement with graph cuts for gene regulatory network inference,BRANE Clust: Cluster-Assisted Gene Regulatory Network Inference Refinement,BRANE page).
As a preparation to a follow-up PhD thesis above subject, the trainee will focus on the statistical integration (Understanding gene regulatory mechanisms by integrating ChIP-seq and RNA-seq data: statistical solutions to biological problems) of different transcriptomic data for multiple microorganisms strains sharing the same genealogy. Feature learning for multilayer networks(Feature Learning in Multi-Layer Networks (OhmNet)) and the exploration of the resulting higher-order networks (Higher-order organization of complex networks) will evaluated in the context on this study.
(c) GdR 720 ISIS - CNRS - 2011-2018.