Informations générales
Intitulé de l'offre : PhD (M/F) in development of Foundation Models for single-cell multi-omics data (H/F)
Référence : UMR3738-DEBPHI-005
Nombre de Postes : 1
Lieu de travail : PARIS 15
Date de publication : jeudi 8 janvier 2026
Type de contrat : CDD Doctorant
Durée du contrat : 36 mois
Date de début de la thèse : 1 septembre 2026
Quotité de travail : Complet
Rémunération : 2300 € gross monthly
Section(s) CN : 51 - Modélisation mathématique, informatique et physique pour les sciences du vivant
Description du sujet de thèse
Single-cell high-throughput sequencing, extracting huge amounts molecular data from a cell, is creating exciting opportunities for machine learning to address outstanding biological questions. The PhD, to be recruited in the context of the ERC StG MULTI-viewCELL, will be working on the development of multi-modal Foundation Models integrating single-cell multi-omics data or spatiotemporal information.
PhD activities :
- design of a new mathematical method
- monitoring and study of publications relevant to the field
- programming/coding in Python (Pytorch)
- presentation of results at conferences
- interaction with team members and international collaborators
Contexte de travail
The Machine Learning for Integrative Genomics team (https://research.pasteur.fr/en/team/machine-learning-for-integrative- genomics/) at Institut Pasteur, led by Laura Cantini, works at the interface of machine learning and biology (tools developed by the team: https://github.com/cantinilab). The team is composed of 9 people : 4PhD students, 3 post doc, 1 research engineer and 1 assistant. The team is associated with the Institut Pasteur's Computational Biology Department, UMR3738 and the PRAIRIE Artificial Intelligence Institute. The team recently won ERC StG funding, which is the subject of this recruitment.
MULTIview-CELL ERC StG :
https://www.ins2i.cnrs.fr/fr/cnrsinfo/laura-cantini-un-projet-erc-starting-grant-linterface-entre-apprentissage-automatique-et
https://research.pasteur.fr/en/project/multi-viewcell/
Contraintes et risques
Work on computer
Informations complémentaires
We expect a candidate with a strong background in machine learning or statistics. The candidate must also be proficient in high-level languages like Python. Familiarity with single-cell date and experience with existing single-cell methods and software would represent a strong advantage. Excellent communication skills and team spirit, and an ability to work in autonomy are essential. Fluent English both spoken and written is required.
Required degree : Master of data science, computer science, applied mathematics