진학프로석·박사 채용 접수 플랫폼

Work environment :

The team is based at the Laboratory of Pathogens and Host Immunity (LPHI) at the University of Montpellier and focuses on applying mathematical and computational approaches to biological systems. Its core activity is the development of deep learning methods for protein design and optimization, with applications in biology and medicine. The research is highly interdisciplinary and relies on the integration of experimental and theoretical research.

The position: We are seeking a motivated postdoc with a background in biophysics/computer science/bioinformatics, and a strong interest in machine learning and protein design. The project will focus on developing a novel SRP designing pipeline that integrates physical modeling into deep learning architectures.

Main Mission :

Currently, SRPs design relies heavily on expensive and time-consuming experiments. The reason why the omnipresent AI methods have not yet been exploited in the SRP-design field can be attributed to a typical characteristic of modern deep learning algorithms: they are very parameter-rich, therefore requiring large amounts of data for effective training.

To overcome the limitations of data scarcity in training deep learning models, one effective strategy is to constrain the model using known physical principles. Since SRPs are protein-based structures, their behavior can be guided by established rules of protein biophysics. Incorporating these principles directly into the model reduces the challenge of learning from limited data and improves robustness. The final goal of the project is to design a computational pipeline that integrates physical principles and data driven DL methods to automatically design SRPs to be used as drug carriers. Such a framework could be especially valuable in treating conditions like cancer metastasis, where locating and reaching affected tissues remains a major challenge.

Activities :

The discovery of new therapeutic agents has significantly improved medical treatments, but their effectiveness depends on precise delivery to the target site. Stimuli-responsive polypeptides (SRPs) are protein-based drug carriers that offer a high degree of customization and can be programmed to specifically release a drug when encountering a specific chemical microenvironment. Unlike conventional systems made from organic materials or polymers, which often lack the flexibility needed to function in complex and variable disease settings, SRPs can be tailored to recognize a wide range of physical and chemical triggers. From the molecular point of view, many SRPs have been shown to release the drug via a reversible liquid liquid phase separation (LLPS). However, like it happens for many other proteins, small changes in their sequences can lead to significant differences in their emergent properties, such as their reaction to a stimulus. This characteristic makes them versatile but also challenging to design and control, due to the absence of adequate mathematical models.

Contract duration : 17 months

Salary : 2,271€ per month