M2 - Development in python of a method for reconstruction and prediction of metabolic networks to analyse the metabolic impact of the MDM2 gene on cancer cells

Type de poste
Niveau d'étude minimal
Dates
Durée du poste
Contrat renouvelable
Contrat non renouvelable
Date de prise de fonction
Date de fin de validité de l'annonce
Localisation
Nom de la structure d'accueil
Adresse

UMR 1331 TOXALIM INRA / ENVT / INP EI Purpan / UPS
180, chemin de Tournefeuille
31027 TOULOUSE 3
France

Contacts
Pablo RODRIGUEZ
Nathalie POUPIN
Email du/des contacts
pablo.rodriguez-mier@inrae.fr
nathalie.poupin@inrae.fr
Description

Context

Metabolism and its regulation are a set of complex and closely coordinated processes involving hundreds, even thousands of enzymes, reactions, metabolites and genes, whose interactions define complex networks that are unique to each species. This complexity gives organisms the flexibility to adapt their energy functions and growth requirements to a wide variety of conditions. The malfunctioning of these mechanisms plays a central role in the development of many diseases, but particularly in cancer, where cancer cells exploit metabolic reprogramming to their own advantage to maintain a rapid rate of proliferation and survive under conditions of hypoxia, nutrient depletion and even develop resistance to treatment. Being able to accurately detect these metabolic changes or deregulations would be beneficial not only for a better understanding of biological systems, but also for the development of more targeted therapies and treatments for many diseases.

One way to detect significant changes in metabolic functions is to use calculation techniques based on metabolic networks. Metabolic network reconstruction is a technique that provides comprehensive and functional information on the potential metabolic states of a cell from experimental data such as gene expression. This type of technique integrates experimentally measured gene expression data into global metabolic networks (several thousand reactions) to predict which reactions are potentially active in the studied state. To do this, the objective is to obtain the metabolism sub-network that is most consistent with the experimental data, while respecting the physico-chemical constraints of metabolism. This enables us to predict the most probable metabolic state of the cells.

The most commonly used techniques focus on obtaining a single metabolic network that addresses this problem, whereas in fact there are many possible solutions that explain the experimental data. It is important to know this space of possible metabolic networks in order to better interpret the metabolic state. In the team, we are working on the development of a method allowing the enumeration of this set of possible metabolic networks on the basis of transcriptomics data.

Objectives

The main objective is the implementation of a Python algorithm for the reconstruction and enumeration of metabolic networks and its application to a data set to study the effects of the MDM2 gene on the metabolism of cancer cells. These transcriptomic data have been obtained and are available within the framework of a collaboration with the Cancer Research Institute of Montpellier (Laetitia Linares' team).

This project will require the lifting of several methodological and technological barriers:

  • Learn the basics of modelling and optimisation under constraints for the calculation and simulation of metabolic flows.

  • Know and use the open source Cobrapy python library.

  • To understand optimisation problems for the reconstruction of metabolic networks.

  • Programming new techniques to generate multiple solutions with Python, using the Optlang library with Cobrapy.

  • To understand and interpret real biological data on the metabolic function of the MDM2 gene, important in cancer.

  • Ensure the continuity of the code (versioning, documentation, tests)

 

Desired competences

 

The candidate will have skills in bioinformatics and mathematics:

  • Python (+++), Matlab (+)

  • Linear algebra, optimisation (+)

  • Graphs (+), notions on metabolic networks (+)

  • Interest in metabolism and physiology

Supervision

The internship will be carried out within a team of bioinformaticians in collaboration with biologists who are experts in the study of cancer cell metabolism. It will be supervised by Pablo RODRIGUEZ, INRAE postdoc (INRAE, UMR1331 Toxalim) and Nathalie Poupin (INRAE Research Fellow, UMR1331 Toxalim).

Required information

Interested students should send their CV and a letter of motivation before 15 October explaining why they are interested in the project to pablo.rodriguez-mier@inrae.fr

Equipe adhérente personne morale SFBI
Equipe Non adhérente