GlycoDENSITY is an open-source program that refines GlycoSHIELD predictions using experimental (volumetric) data, such as cryoEM, cryoET maps, or X-ray electron density maps. GlycoDENSITY sorts glycan conformers grafted with GlycoSHIELD according to their fit to the underlying experimental data.


GlycoDENSITY is released under a specific license and is available for download here. By downloading the code you acknowledge having read and accepted the terms of the license.

Use case

GlycoDENSITY GlycoSHIELD ensemble refinement. Here we used SARS-CoV-2 Spike protein in its post-fusion state (red). Associated map contains low resolution information about glycans (e.g. marked in magenta box), which is only allowing to trace first monosaccharides of a glycan (first image in the bottom row, density is shown as outline, glycan as sticks). The remaining low resolution density can be used to refine GlycoSHIELD ensemble. Glycans are first grafted with GlycoSHIELD and a synthetic density map is generated to visualise span of the glycan mushroom. Next, GlycoDENSITY is used to select a small number of best-fit conformers (green, blue, brown, red). Finally, we can generate a synthetic density map based on these conformers and see how well it represents the experimental density.