Postdoc M/F Physics-Informed AI for High-Dimensional Modeling of the Tumor Microenvironment

Job Description

Missions

Mathematical modeling and numerical methods
Code development
Presentation of results in peer-reviewed journals and conferences

Activity

The postdoc will join the SIMAI project, funded by INSERM's MIC program Interdisciplinary Approaches in Oncogenic Processes and Therapeutic
Perspectives: Contributions of Mathematics and Informatics to Oncology. It investigates the relationship between physical modeling and high-dimensional data,
an active and rapidly evolving area in artificial intelligence and machine learning.
The theoretical approaches will benefit from spatial proteomics and transcriptomics data provided by our biology partners at IRCM, the Montpellier Institute for Cancer Research. Our goal is to predict the response of melanoma to immune checkpoint inhibitor (ICI) treatment.
As part of the SIMAI project, multi-omic data will be used to construct 3D models of the tumor microenvironment (TME), which represents a complex ecosystem of dynamically interacting cells of various types. A deeper understanding of this ecosystem could yield critical insights into disease progression, particularly the emergence of resistance to treatment. This project aims to synergize artificial intelligence (AI) with mechanistic models of tumor heterogeneity,
combining the strengths of both approaches.
The project will build on recent work from Radulescu's lab in multidimensional and multimodal reconstruction of the TME
(Hodgkinson et al., 2022; Arslan et al., 2023; Kumar et al., 2024). For mechanistic modeling, we use a mesoscale modeling framework based on partial differential equations (PDEs), which integrates population dynamics with intracellular signaling, metabolism, and gene regulation into a unified formalism. This approach models cell densities over time, space, and internal (structural) dimensions (Hodgkinson et al., 2022).

In mesoscale tumor modeling, cell behavior is often governed by high-dimensional partial differential equations, in which cells are characterized not only by their spatial location but also by internal variables representing signaling and metabolic pathway states (Hodgkinson et al 2022). This high dimensionality poses significant challenges for conventional numerical methods, such as finite difference or finite element schemes, due to the rapid growth in computational
cost and associated limitations in stability and accuracy. The postdoc will investigate the use of physics-informed neural networks (PINNs) as a mesh-free, collocation-based approach for solving such high-dimensional PDEs, avoiding explicit spatial discretization grids. For parameter inference from experimental data, the postdoc could employ physics-informed generative adversarial networks (PI-GANs), which integrate observational data with governing physical constraints.

Your Profil

Skills

Strong theoretical background in partial differential equations,
Familiarity with artificial intelligence techniques,
Proficiency in Python, Julia, MATLAB.
PhD degree in applied mathematics, computer science, or a related discipline.

Your Work Environment

CSB in LPHI is a leading academic team specializing in multiscale modeling of biological systems. The group combines expertise in mathematical modeling, simulation of cellular and tissue dynamics, machine learning approaches for predictive biology, with biological and clinical expertise. Secure data storage and computational infrastructure are available at both the team and institutional levels. At the team level, this includes a dedicated subnet with SSH bastion access, LDAP-controlled permissions, and computations running on virtual machines for isolation and resource management. At the institutional level, UM offers secure, encrypted, and auditable storage fully compliant with medical data regulations (GDPR, HIPAA), as well as high-performance computational resources, including NVIDIA H100 GPUs.

Compensation and benefits

Compensation

from 3071€ gross per month depending on experience

Annual leave and RTT

44 jours

Remote Working practice and compensation

Pratique et indemnisation du TT

Transport

Prise en charge à 75% du coût et forfait mobilité durable jusqu’à 300€

About the offer

About the CNRS

The CNRS is a major player in fundamental research on a global scale. The CNRS is the only French organization active in all scientific fields. Its unique position as a multi-specialist allows it to bring together different disciplines to address the most important challenges of the contemporary world, in connection with the actors of change.

CNRS

The research professions