Team: Tancrède Salotti (Master 2 at Sorbonne University & EPITA), Aria Gheeraert (INSA Toulouse), Yasser Mohseni (DSIMB, Université Paris Cité)

Protein-carbohydrate interactions play crucial role in various biological processes. However, such interactions remain poorly investigated due to the difficulty of their experimental description. The main goal of the project is accurate classification and development of machine learning based tools for prediction of potential carbohydrate binding sites on the protein surface. This project has received a young researcher grant by French National Research Agency (ANR) under the title "SugarPred: Deciphering protein-carbohydrate interactions" and Master Student Scholarship funded by Data Intelligence Institute of Paris. 

References:

1. Gheeraert A, Guyon F, Pérez S, Galochkina T. Unraveling the diversity of protein-carbohydrate interfaces: insights from a multi-scale study. Carbohydr Res, 190377 (2025). DOI: 10.1016/j.carres.2025.109377
2. Gheeraert A, Bailly T, Ren Y, Hamraoui A, Te J, Vander Meersche Y, Cretin G, Leon Foun Lin R, Gelly J-C, Pérez S, Guyon Frédéric & Galochkina T. DIONYSUS: a database of protein-carbohydrate interfaces.  Nucleic Acids Res 53(D1), D387-D395 (2025). DOI: 10.1093/nar/gkae890 

Team: Yann Vander Meersche (Whitelab Genomics), Thiziri Chennit (Master 2 ISDD, Université Paris Cité), Gabriel Cretin (DSIMB, Université Paris Cité), Jean-Christophe Gelly (DSIMB, Université Paris Cité)

Protein dynamic properties determine protein function. However, their experimental description is material and time consuming and therefore not convenient for large scale studies. In the current project we aim to fill this gap by developing machine learning tools for prediction of global and local deformability of protein regions from their sequence/structure. The first results for protein flexibility prediction in terms of expected B-factors were published in the Journal of Molecular Biology and are available in the form of an open-source web server MEDUSA.  Local conformations of a given protein can be predicted via the PYTHIA tool, recently developed by our team. Our current major contribution in the field is the development of ATLAS database of molecular dynamics simulations extensively used by the community for the development and benchmarking of the deep learning-based tools. In particular, our team has released PEGASUS for fast prediction of protein flexibility profiles directly from amino acid sequences. 

References:

1.  Vander Meersche Y, Cretin G, Gheeraert A, Gelly J-C, Galochkina T. ATLAS: protein flexibility description from atomistic molecular dynamics simulations. Nucleic Acids Res 52(D1), D384–D392 (2024). DOI: 10.1093/nar/gkad1084
2.  Vander Meersche Y, Diharce J, Gelly J-C, Galochkina T. Flexibility or uncertainty? A critical assessment of AlphaFold 2 pLDDT. Structure (2025). DOI: 10.1016/j.str.2025.09.001
3. Vander Meersche Y, Duval G, Cretin G, Gheeraert A, Gelly J-C, Galochkina T. PEGASUS: Prediction of MD-derived protein flexibility from sequence. Protein Science 34:e70221 (2025). DOI: 10.1002/pro.70221
4. Vander Meersche Y, Cretin G, de Brevern A G, Gelly J-C, Galochkina T. MEDUSA: Prediction of Protein Flexibility from Sequence, J Mol Biol 433(11), 166882 (2021). DOI: 10.1016/j.jmb.2021.166882 
5. Cretin G, Galochkina T, de Brevern A G, Gelly J-C. PYTHIA: Deep Learning Approach for Local Protein Conformation Prediction, IJMS 22(16), 8831 (2021). DOI: 10.3390/ijms22168831 

During my PhD I was developing molecular models of bacterial membranes and investigating conformational behavior of their components (lipopolysaccharides) in different environment in order to better understand bacterial membrane organization. As a post-doc, I have studied molecular mechanics of the human glucose transporter GluT1 during glucose transfer across the membrane. Now I continue working on models of different membrane systems with experimental collaborators (as can be seen in our recent paper on ABCG2 transporter) in order to better understand the impact of different factors on membrane protein dynamics.

References:

1. Buks R, Brusson M, Cochet S, Galochkina T, ..., El Nemer W. ABCG2 Is Overexpressed on Red Blood Cells in Ph-Negative Myeloproliferative Neoplasms and Potentiates Ruxolitinib-Induced Apoptosis, IJMS 22(7), 3530 (2021). DOI: 10.3390/ijms22073530 
2. Galochkina T, Chong M N F, Challali L, Abbar S, Etchebest C. New insights into GluT1 mechanics during glucose transfer, Sci Rep 9, 998 (2019). DOI: 10.1038/s41598-018-37367-z 
3. Galochkina T, Zlenko D, Nesterenko A, Kovalenko I, Strakhovskaya M, Averyanov A, Rubin, A. Conformational dynamics of the single lipopolysaccharide O-antigen in solution, ChemPhysChem 17(18), 2839-53 (2016). DOI: 10.1002/cphc.201600524