Mots-Clés
Deep learning
protein design
TCR
Immunotherapy
Description
Deep Learning–Based Design of T-Cell Receptors
Project description
T-cell receptors (TCRs) play a central role in adaptive immunity by recognizing antigenic peptides presented by the major histocompatibility complex (MHC). The specificity and strength of this interaction are largely determined by the complementarity-determining regions (CDRs) of TCR, particularly the hypervariable CDR3 loops. Rational design of CDRs with enhanced binding affinity has profound implications for immunotherapy, vaccine development, and autoimmune disease research.
Recent advances in deep learning have shown potential for modeling protein complexes, protein design, and extracting contextual information from protein sequences. However, despite these advances at the interface of deep learning and computational biology, designing CDRs that achieve high binding affinity remains a significant challenge.
The objective of this internship is to benchmark and develop deep learning–based methods for designing CDRs of TCR with optimized binding affinity and structural stability, advancing computational tools for therapeutic immunoengineering. The internship is scheduled to take place between January and June 2026 under the supervision of Jean-Christophe Gelly and Yasser Mohseni Behbahani. It will be hosted by the Dynamics of Structures and Interactions of Macromolecules in Biology (DSIMB) team within the Integrated Biology of the Red Blood Cell (BIGR) lab, located at Hôpital Necker, 75015 Paris, France.
Research context
The internship will be divided into two main phases:
* Benchmarking existing methods, such as the RFdiffusion family of models, for CDR design.
* Developing, training, and assessing a structure-based model capable of predicting and optimizing CDR sequences and structures to improve TCR–pMHC (peptide–MHC) binding.
The project will integrate sequence representation learning (via protein language models) and structural representation learning (via graph neural networks) towards a design process guided by structural and biophysical priors. The specific objectives include:
* Collect and preprocess publicly available TCR–pMHC complex datasets, including both sequence and structural information.
* Evaluate and compare existing deep learning-based models in terms of generative and predictive power.
* Design and train an in-silico, structure-based model to generate optimized CDR variants predicted to maximize both binding affinity and structural stability.
* Validate model predictions across multiple applications, including inflammatory and autoimmune diseases.
Benefits
* 6-month paid internship (stipend provided).
* Access to high-performance computing resources.
* Opportunity to collaborate with an interdisciplinary team of experts in computational biology and deep learning.
* Subsidized meals.
Candidate profile
* Applicants must be enrolled in a Master’s program in Computer Science, Bioinformatics, or a related field.
* Strong programming skills in Python and solid coding practices.
* Practical experience with deep learning frameworks (e.g., PyTorch or TensorFlow).
* Good knowledge of computational biology.
* Ability to work effectively in a collaborative, interdisciplinary environment.
* Strong written and oral communication skills in English.