Mots-Clés
peptides
membranes
deep learning
antibiotics
biophysics
simulations
AI
drug design
Description
Project Context
Join a cutting-edge project at the intersection of computational chemistry, biophysics, and machine learning to address one of the most pressing global health challenges: antimicrobial resistance. The project, funded by the MAIA initiative (https://maia.a2u.fr), aims to develop predictive models for the membrane permeability of antimicrobial peptides (AMPs) and cell-penetrating peptides (CPPs)—key candidates for next-generation antibiotics and drug delivery systems.
Despite their therapeutic potential, AMPs/CPPs rarely reach clinical trials due to challenges in evaluating their membrane affinity and permeability, which are critical for efficacy and toxicity. This project will leverage COSMOperm (a thermodynamic model for membrane permeability) and graph neural networks (GNNs) to create a high-throughput screening tool. The successful candidate will use and contribute to ADAPTABLE, a web platform hosting >40,000 AMPs, and collaborate with experts in quantum chemistry (LG2A) and AI-driven molecular modeling (GEC).
Key Responsibilities
• Data Generation: Use COSMOperm to compute thermodynamic properties for small oligopeptides
• Model Development: Encode peptides as graphs, implement graph convolutional networks
• Model Training/Validation: Optimize hyperparameters and train models to predict the permeability of linear and cyclic AMPs/CPPs.
Required Qualifications
• PhD in Computational Chemistry, Bioinformatics, Machine Learning, or a related field.
• Strong programming skills in Python (PyTorch, PyG, or similar libraries).
• Experience with machine learning (deep learning, graph-based methods, or molecular modeling is a plus).
• Familiarity with Linux/GPU computing and data analysis tools.
• Proactive and collaborative mindset—ability to bridge chemistry and AI disciplines.
Desirable:
• Knowledge of thermodynamic models (COSMO-RS/COSMOperm) or peptide biophysics.
• Experience with molecular modeling and quantum chemistry.
Work Environment
• GEC Laboratory (CNRS/UPJV): A dynamic team with expertise in antimicrobial peptides and AI applied to molecular modeling.
• LG2A Laboratory (UPJV): Collaborative access to COSMOperm and quantum chemistry resources.
• Resources: High-performance computing (GPUs), interdisciplinary network (MAIA Chemistry/Health axis), and potential for publication in high-impact journals.
Application Process
Interested candidates should send the following to benjamin.bouvier@u-picardie.fr:
1. CV (including publication list).
2. Cover letter (1–2 pages) detailing:
◦ Your motivation for the project.
◦ Relevant experience in machine learning or molecular modeling.
◦ Preferred start date.
3. Names/contact details of 2–3 references.
Review of applications will begin immediately and continue until the position is filled.
Why Apply?
• Impact: Contribute to global health by accelerating the discovery of antimicrobial therapies.
• Innovation: Develop novel AI tools for drug design, with potential for patents/publications.
• Network: Collaborate with MAIA’s interdisciplinary community (chemistry, health, AI).
Related Papers:
• Bouvier, J. Chem. Theory Comput 2026, 22, 1215.
• Ramos-Martin et al, Life Sci. Alliance 2019, 2, e201900512.
• Leal et al, BBA Biomembranes 2026, 1868, 184525.