Date : 2026-09-27 => 2026-09-29
Lieu :
URL: https://bbb.unistra.fr/rooms/gil-kgo-eqc-4z0/join
Pass code: 365251
Dear colleagues,
The fourth lecture in the series of advanced chemoinformatics is dedicated to Mrs Mireille KRIER from Cadence / OpenEye company.
Title – Shape and color: A unifying principle for modelling molecules
Presenter: Mrs Mireille KRIER from Cadence / OpenEye company
Date: 29/09/2026 at 6pm (Strasbourg time)
Molecular shape and electrostatics are fundamental determinants of molecular recognition and protein-ligand binding. This presentation introduces a unified computational framework in which molecules are represented by three-dimensional shape together with discrete chemical “color” features. Spherical Gaussian functions provide an efficient and differentiable approximation of molecular volume and chemical features, enabling rapid calculation and optimization of molecular overlay.
This shape-and-color representation underpins a range of ligand- and structure-based drug discovery methods. ROCS and FastROCS perform molecular alignment and similarity searching across collections ranging from conventional compound databases to billion-molecule libraries. The same principles support fragment replacement with BROOD, protein binding-site comparison with SiteHopper, and protein-ligand complementarity assessment with FRED and HYBRID. Flexible approaches such as ShapeFit, FlexiROCS, and AFITT extend the framework to pose prediction and fitting against experimental density.
Aligned shape-and-color representations can also serve as descriptors for interpretable three-dimensional quantitative structure-activity relationship models. The resulting models capture spatial features associated with binding affinity and can be transformed into queries for prospective virtual screening. Finally, ROCS X extends three-dimensional similarity searching to synthon-based chemical spaces. Collectively, these applications demonstrate how shape and color provide a consistent and computationally efficient foundation for molecular modelling, virtual screening, docking, lead optimization, and predictive modelling.
URL: https://bbb.unistra.fr/rooms/gil-kgo-eqc-4z0/join
Pass code: 365251
This lecture is part of the curriculum for the Master’s program in Chemoinformatics. Students enrolled in this program will be required to take a test on the lecture following it.
With the support of the Erasmus Mundus project « Chemoinformatics+ », Société Française de Chémoinformatique, Chaire Industrielle ChIMIA, ECPM/Chimie&IA, GdR ThéMoSiA, BigDataChim and MADICS/DSChem.
Best regards,
Gilles Marcou
Notre site web : www.madics.fr
Suivez-nous sur Tweeter : @GDR_MADICS
Pour vous désabonner de la liste, suivre ce lien.