Vous êtes ici : Accueil » Kiosque » Annonce

Identification

Identifiant: 
Mot de passe : 

Mot de passe oublié ?
Détails d'identification oubliés ?

Annonce

11 octobre 2019

Finalize an image processing pipeline and generate a database for statistical analysis


Catégorie : Ingénieur


3 months contract Research Assistant

Objective : Finalize an image processing pipeline and generate a database for statistical analysis
Scientific environment : collaborative Inria/I3S/iBV Morpheme Team, https://team.inria.fr/morpheme/
Working place : INRIA/I3S campus
Profile required : computer science student M1-M2, python programming, statistics
Keywords : Software architecture, Database manipulation (SQL), IHM

 

I. Context
The precise spatio-temporal control of gene expression relies on the assembly of macromolecular complexes
called RNP granules containing various protein and RNA molecules (Besse and Ephrussi 2008, Medioni et al
2012, De Graeve and Besse 2018). Currently, it is still not clear how these RNP granules, visible in light microscopy
are assembled and regulated within the cell. The laboratory of F. Besse has shown that the Imp
RNA binding protein accumulates as RNP granules in the cytoplasm of cultured cells. Remarkably, mutant
forms of Imp unable to bind RNA do not assemble into RNP granules while a mutant form devoid of its Cterminal
domain generates bigger granules. These results indicate that i) Imp granule properties are finely
regulated, and ii) quantitative analyses can be performed.
To systematically identify the genes regulating Imp+ RNP granules,
we have performed a high-throughput microscopy screen that
generated 3 millions of images, each containing dozens of cells (Fig.1).
Our objective is now to develop an automatic image analysis pipeline
for extraction of RNP granule features in all tested conditions. Individual
RNP granule features will be stored in a database that will be
queried for statistical analyses and identification of candidate regulatory
genes. So far, we have already developed algorithms to identify healthy
cells and accurately detect RNP granules and we are finalizing an
algorithm to segment individual cells. These tools will serve as
building blocks for the construction of the fully automatic pipeline.


II. Objective of the internship
We are now looking for a dynamic, interactive and highly motivated candidate who will :
- develop a « user-friendly » pipeline that can be made available to the community of cell biologists
- make a web page for end-user
- run the pipeline on the collection of images and store granule features in an tailored database
- perform basic statistical analyses to identify interesting hits (this includes statistical moments for number
and size of granules and simple statistical tests for measuring clustering effect).
The candidate will be closely supervised by both computational scientists (X. Descombes, E. Debreuve) and
biologists (F. De Graeve, F. Besse).
Interested candidates should send their CV and a short motivation letter to fabienne.de-graeve@univ-cotedazur.fr.


III. Communications related to the project
2016, De Graeve F, Kozlowski D, Cedilnik N, Debreuve E, Descombes X, Besse F. Drosophila Imp granules,
a paradigm to identify factors that regulate the assembly, stability and distribution of RNP granules
The multiple facets of RNA in development and disease, Nice, France.
2018, Nicolas Cedilnik, Eric Debreuve, Fabienne de Graeve, Florence Besse, Xavier Descombes. SPADE: A
Small Particle Detection Method Using A Dictionary Of Shapes Within The Marked Point Process Framework.
IEEE International Symposium on Biomedical Imaging (ISBI) 2018 (2018-04-04)
2018, De Graeve F, Rahmoun S, Kozlowski D, Cedilnik N, Debreuve E, Descombes X, Besse F. An image
based high throughput screen to identify regulators of Imp containing RNP granules. 32nd Ann. Conf. of
French Drosophilists, Hyeres, France.
2019 Fabienne De Graeve, Eric Debreuve, Somia Rahmoun, Szilvia Ecsedi, Alia Bari, Arnaud Hubstenberger,
Xavier Descombes, Florence Besse Detecting and quantifying stress granules in tissues of multicellular
organisms with the Obj.MPP analysis tool Traffic. 2019 Sep;20(9):697-711

 

Dans cette rubrique

(c) GdR 720 ISIS - CNRS - 2011-2020.