As the prospects of revolutionizing medicine through targeted drug delivery become more fully realized, the demand for safe, sustainable, and effective drug carrier materials has risen commensurately. In the field of nanomedicine, glycans have long been highly desirable for use in targeted drug carrier materials as they are known to display receptor specificity with antibody-like precision while possessing far greater biosafety and immunomodulatory properties. Despite this, the complex and highly dynamic structural nature of glycans often presents a significant hindrance to traditional experimental methods of study and design. The past two decades have shown, however, that computational simulation of glycosylated systems can overcome much of what experiments lack in capability and are even becoming a gold-standard in the design of smart drug delivery systems. While computational simulation and design of glycosylated drug delivery systems has seen some advancements in recent years, a comprehensive methodology capable of accommodating the extensive design space inherent to glycomaterials has not yet been fully established. It is here that we propose a computational platform equipped to efficiently model, characterize, and simulate both observed and novel glycan structures towards their actualization as a targeted biomaterial for use in nanomedicine. Glycan modeling tools such as RosettaCarbohydrate and the CHARM36 force field parameters provide robust cheminformatic support for precise replication of mechanistic phenomena inherent to complex carbohydrates in silico, while glycan characterization is performed through high-throughput molecular docking simulation between constructed glycoligand candidates and PDB-tabulated carbohydrate-binding receptors relevant to the drug delivery field such as lectins, antibodies, and certain viral proteins. Inspired by the process of lead optimization for structure-based drug design, our modular computational pipeline employs a fragment-based approach to structural screening and characterization tailored to the unique physiochemical properties of glycans; this is achieved through docking algorithms like Rosetta’s GlycanDock and Scripps’ Vina-Carb which allows for exhaustive sampling of the complex torsional angles involved in glycosidic linkages which are often not sufficiently represented by other docking suites such as Vina and Schrodinger Glide. Additionally, by sampling oligomeric permutations of carbohydrates replicating common glycan motifs documented in literature, useful parameters such as binding affinity, interface score, and residue contacts can be tabulated for a multitude of glycan-fragments cross-docked with a large repository of potential receptor targets. To ensure high-resolution thermodynamic data are obtained, high-scoring docked poses predicted by GlycanDock are further explored through Schrodinger IFD-MD to consider receptor flexibility and estimate structural stability over time. Prospects for later features of this analysis and design pipeline include support for GlycanDock’s “blind docking” which shows success in docking glycoligands for which the receptor binding site is unknown, and incorporation of GROMACS pose clustering to explore extended binding mode sampling beyond that of the top scoring pose. This work will establish the foundation for a modular computational pipeline aiding in the design of future drug delivery systems incorporating glycan-promoted target specificity. Additionally, this investigation will establish a comprehensive database of predicted binding affinity between common glycan residue motifs and carbohydrate binding proteins found throughout the body as potential receptor targets for future drugs.
