It is often difficult to be certain which genes underlie the effects seen in association studies. However, variants that disrupt the protein, such as predicted loss of function (pLoF) and missense variants, provide a shortcut to identify genes with a clear biological link to the phenotype of interest. Glycosylation is one of the most common post-translationalmodifications of proteins, and an important biomarker of both disease and its progression. Here, we utilised the power of genetic isolates, gene-based aggregation tests and intermediate phenotypes to assess the effect of rare (MAF<5%) pLoF and missense variants from whole exome sequencing on the N-glycome of plasma transferrin (N=1907) and immunoglobulin G (N=4912), and their effect on diseases. We identified significant gene-based associations for transferrin glycosylation at 5 genes (p<8.06x10-8) and for IgG glycan traits at 4 genes (p<1.19x10-7). Associations in three of these genes (FUT8, MGAT3 and RFXAP) are driven by multiple rare variants simultaneously contributing to protein glycosylation. Association at ST6GAL1, with a 300-fold up-drifted variant in the Orkney Islands, was detectable by a single-point exome-wide association analysis. Glycome-associated aggregate associations are located in genes already known to have a biological link to protein glycosylation (FUT6, FUT8 for transferrin; FUT8, MGAT3 and ST6GAL1 for IgG) but also in genes which have not been previously reported (e.g. RFXAP for IgG). To assess the potential impact of rare variants associated with glycosylation on other traits, we queried public repositories of gene-based tests, discovering a potential connection between transferrin glycosylation, MSR1, galectin-3, insulin-like growth factor 1 and diabetes. However, the exact mechanism behind these connections requires further elucidation.