Séminaire: Marie-Pierre Gaigeot, Professeure à l'Université d'Evry Val d'Essonne

I will review some of our developments of the past years in algorithmic graph theory and their applications in chemo-informatics. I will review the definitions of our topological molecular graphs labeled 2D-MolGraphs1 at the atomic level of representation that make them transferable to represent gas phase molecules and clusters, liquids, and more complex inhomogeneous interfaces between solid and liquid just as easily, without modifications. 2D-MolGraphs and associated graph algorithms have been applied to the automatic analysis of the conformational dynamics arising in molecular dynamics (MD) simulations (ab initio DFT-MD and classical FF-MD) of complex molecular systems. I will show a few examples of our achievements extracted from some of our recent works, see e.g. reviews2-4, for instance for the time evolution of peptidic chains in the gas phase, for the specific hydrophobic molecular organization of liquid water (2D-HB-Network) at the interface with hydrophobic media (air, graphene, etc), and how these graphs can help to understand and rationalize freezing processes at alcoholic air-water interfaces.5 2D-MolGraphs have also been included in workflows for the design of catalysts’ molecules in homogeneous catalysis.7-8 We will show how the 2D-MolGraphs can hence easily be included in automated high-throughput in silico reactivity workflows, and how essential they are for some of the decisive steps to be taken in these workflows.

These topological 2D-MolGraphs can be coarse-grained in order to represent molecules and molecular assemblies at a larger coarse grain level. This has been achieved in ref.6 with the development of coarse-grained polygraphs of H-Bonded cycles typically applied to peptides which molecular structure is based on H-Bonded cycles/polygons. We will show how the time-evolution of these polygraphs help to rationalize the conformational dynamics of peptides at various temperatures. Key-elements in these developments are algorithms of isomorphism of the 2D-MolGraphs and of polymorphism of the coarse-grained H-Bonded cycles, that allow to reveal meta-structures that cannot be seen otherwise at the atomic level of representation.
If time permits, I will also briefly outline our current efforts in the implementation of machine learning algorithms for specific tasks associated to graphs and to vibrational spectroscopies, with the purpose of the inverse design of molecules and materials from spectroscopy.
These works have been/are conducted in collaboration with: Dr Sana Bougueroua (computer scientist),1 Dr Alvaro Cimas (theoretical chemist),1 Prof Dominique Barth (computer scientist),2 Prof Jérémie Cabessa (computer scientist),2 Dr Ylène Aboulfath (computer scientist),2 Dr Ali Hashemi (computer scientist & theoretical chemist),3 Prof Evgeny Pidko (theoretical chemist)3
1 Université Paris-Saclay, Univ Evry, CY Cergy Paris Université, CNRS, LAMBE, 91025 Evry-Courcouronnes, France
2 Université Paris-Saclay, Univ Versailles Saint Quentin, DAVID, 78035 Versailles, France
3 Inorganic Systems Engineering, Department of Chemical Engineering, Faculty of Applied Sciences, Delft University of Technology, 2629 HZ Delft, The Netherlands

References :

1- S. Bougueroua, R. Spezia, S. Pezzotti, S. Vial, F. Quessette, D. Barth, M.-P. Gaigeot
Graph theory for automatic structural recognition in molecular dynamics simulations.
J. Chem. Phys. 149:184102-15 (2018)

2- S. Bougueroua,Y. Aboulfath, A. Cimas, A. Hashemi, E.A. Pidko, D. Barth, M.-P. Gaigeot,
Topological graphs: a review of some of our achievements and perspectives in physical-chemistry and homogeneous catalysis
In Press, Comptes Rendus. Chimie (2024)
DOI: 10.5802/crchim.317

3- S. Bougueroua, M. Bricage, Y. Aboulfath, D. Barth, M.-P. Gaigeot
Algorithmic Graph Theory, Reinforcement Learning and Game Theory in MD Simulations: From 3D Structures to Topological 2D-Molecular Graphs (2D-MolGraphs) and Vice Versa
Molecules 28 : 2892-2912 (2023)
DOI: 10.3390/molecules28072892

4- S. Bougueroua, Y. Aboulfath, D. Barth, M.-P. Gaigeot
Algorithmic graph theory for post-processing molecular dynamics trajectories
Mol. Phys. 121 : e2162456 (2023)
DOI: 10.1080/00268976.2022.2162456

5- R. AbouHaidar, S. Bougueroua, D. Duflot, M.-P. Gaigeot, B. Wyslouzil, C. Toubin
Unraveling Aqueous Alcohol Freezing : new theoretical tools from graph theory to extract molecular processes in MD simulations
In Press, Faraday Discussions ‘Atmospheric chemistry in cold environments’, 2024
DOI: 10.1039/D4FD00165F

6- Y. Aboulfath, S. Bougueroua, A. Cimas, V. Chantitch, D. Barth, M.-P. Gaigeot
Time-resolved graph of cycles for the polymorphic identification of the H-Bonded network in flexible (bio-)molecules
J. Chem. Theory Comput. 20:1019-1035 (2024)
DOI: 10.1021/acs.jctc.3c01031  

7- A. Hashemi, S. Bougueroua, M.-P. Gaigeot, E.A. Pidko
HiREX: High-Throughput Reactivity Exploration for Extended Databases of Transition Metal Catalysts
J. Chem. Inf. Model. 63: 6081-94 (2023)
DOI: 10.1021/acs.jcim.3c00660

8- A. Hashemi, S. Bougueroua, M.-P. Gaigeot, E.A. Pidko
ReNeGate: A Reaction Network Graph-Theoretical Tool for Automated Mechanistic Studies in Computational Homogeneous Catalysis
J. Chem. Theory Comput. 18 :7470-7482 (2022)
DOI: 10.1021/acs.jctc.2c00404