Maintenance

Due to a planned maintenance at our hosting facility, the SFBI website will be stopped from October 6 to October 8

Postdoc en IA générative pour la génomique

 CDD · Postdoc  · 24 mois    Bac+8 / Doctorat, Grandes Écoles   Structure et Instabilité des génomes · Paris (France)

 Date de prise de poste : Jan. 1, 2026

Mots-Clés

'bioinformatics' 'genome edition' 'artificial intelligence'

Description

Project overview

Duchenne muscular dystrophy (DMD) is a severe X-linked disorder caused by the loss of dystrophin. A promising therapeutic alternative focuses on upregulating utrophin (UTRN)—a functional paralogue that can compensate for dystrophin loss.
This project aims to combine deep learning–based genomic modeling and experimental genome editing to design, predict, and validate targeted DNA modifications that increase utrophin expression in human muscle cells.

The postdoctoral researcher will bridge computational modeling and experimental validation, participating in both the design and interpretation of CRISPR-based editing experiments.

Research plan

  • Year 1 – Computational phase

Integration and curation of multi-omic datasets (RNA-seq, CAGE, histone marks, DNase)

Fine-tuning deep learning models (e.g. Enformer, Borzoi) to predict regulatory activity around the UTRN locus

Development of a generative AI pipeline to propose sequence edits enhancing UTRN transcription

Prioritization of edits based on predicted expression gain and CRISPR feasibility

  • Year 2 – Experimental phase

Collaboration with Dr. Mario Amendola’s wet-lab team to design and test CRISPR-Cas9 constructs

Validation of top-ranked edits in myoblast and iPSC-derived myotube models

Quantification of UTRN upregulation (qPCR, RNA-seq, Western blot, immunocytochemistry)

Iterative refinement of models based on experimental feedback

Candidate profile

We are seeking a highly motivated and creative postdoctoral fellow with expertise in one or more of the following areas:

  • Required skills:

Strong background in computational biology, genomics, or bioinformatics

Proven experience with deep learning or machine learning applied to biological data

Solid programming skills (Python, PyTorch/TensorFlow, data analysis pipelines)

Ability to work collaboratively in an interdisciplinary environment

  • Desirable (not mandatory):

Familiarity with gene regulation, CRISPR design, or epigenomic datasets

Interest in translational genomics and therapeutic genome engineering

Environment

The ARChE team at the Muséum National d’Histoire Naturelle is recognized for its expertise in chromatin structure, genome instability, and AI-driven genomics. The project is embedded in a vibrant interdisciplinary network including Généthon and other partners specialized in muscle biology and gene therapy.

The postdoc will have access to high-performance computing resources, state-of-the-art experimental platforms, and close mentoring by both computational and experimental experts.

Candidature

Procédure : Please send a single PDF including: Cover letter describing your motivation and fit for the project CV with publication list Send applications to: julien.mozziconacci@mnhn.fr

Date limite : Nov. 15, 2025

Contacts

 
 juNOSPAMlien.mozziconacci@mnhn.fr

Offre publiée le Oct. 8, 2025, affichage jusqu'au Nov. 15, 2025