[BBC] Internship – Bayesian methods for disease subtype discovery

Johann Pellet jpellet at eisbm.org
Fri Sep 23 14:55:04 CEST 2016


6-month Programming Internship: France (Lyon)

Employer: CNRS
Location: CNRS-EISBM, Lyon, France
Qualifications: Undergraduate – Bachelor’s degree
Salary: Depending on qualifications and experience, based on CNRS salary grid
Duration: 6 months
Starting date: Available immediately, expected to start by January/February 2017
Key words: Programming, GP-GPU, Python, R, C, Java, Bioconductor, disease sub-types, Systems Biology

An internship position is available as a part of the European Translational Research Information and Knowledge Management Services project (eTRIKS), funded by the Innovative Medicines Initiative (IMI) of the European Commission 7th Framework Programme and the European Federation of Pharmaceutical Industries and Associations (EFPIA).

The selected candidate will be based in Lyon-Gerland within the CNRS-EISBM team led by Charles Auffray, hosted by Claude Bernard University on the Charles Mérieux campus. He will work closely with the team led by Prof Yi-Ke Guo at the Data Science Institute of Imperial College in London, UK.

The aim of the internship is to evaluate the performance of recent algorithms for identification of disease sub-types using Bayesian methods; specifically, this will involve implementing existing programs, where possible exploiting GP-GPU and Apache spark architectures to address the computationally intensive aspect of the problem. It is anticipated that some of the approaches evaluated will be used for analysis of multi omics datasets from new systems medicine projects.

This position allows an intern to participate in a state-of-the-art EU systems biomedicine project.

Required skills: Good programming skills (eg. python, java, C, R); good communication skills in English.

Enthusiastic and motivated candidates are encouraged to contact Prof. Charles Auffray jobs at eisbm.org with a CV and a brief description of their programming skills.




More information about the BBClist mailing list