Glycosylator: a Python framework for the rapid modeling of glycans.

Area of Science:

• Biomolecular modeling
• Glycoinformatics
• Computational biology

Background:

• Carbohydrates, including glycans, are diverse biomolecules crucial for physiological processes.
• N-linked glycosylation, the attachment of carbohydrates to asparagine, is vital in many organisms.
• Existing glycan modeling tools often require specialized software knowledge or are limited to canonical conformations, hindering integration into protein structure workflows.

Purpose of the Study:

• To present Glycosylator, a novel Python framework for glycan identification, modeling, and modification within protein structures.
• To provide a user-friendly tool that integrates glycan modeling into protein structure analysis.
• To offer advanced capabilities for analyzing and modeling complex glycan structures.

Main Methods:

• Developed a Python framework with both an application programming interface (API) and a graphical user interface (GUI).
• Implemented a 2D rendering of glycoproteins for visual inspection of glycosylation.
• Utilized a genetic algorithm for refining glycan conformations and removing steric clashes.
• Incorporated a library of predefined templates for identifying 3D glycans on protein structures.

Main Results:

• Glycosylator successfully generated protein models with glycosylation and without steric clashes.
• The framework allows for the generation of novel complex sugar moieties using the CHARMM force field.
• The software provides comprehensive functionality for glycan analysis and modeling.

Conclusions:

• Glycosylator offers superior functionality for glycan analysis and modeling compared to existing software and webservers.
• Its molecular topology, based on the CHARMM force field, facilitates the creation of new sugar moieties.
• Glycosylator is poised to be a valuable asset for the glycoinformatics and biomolecular modeling communities.