SweetNET: A Bioinformatics Workflow for Glycopeptide MS/MS Spectral Analysis

Abstract

Glycoproteomics has rapidly become an independent analytical platform bridging the fields of glycomics and proteomics to address site-specific protein glycosylation and its impact in biology. Current glycopeptide characterization relies on time-consuming manual interpretations and demands high levels of personal expertise. Efficient data interpretation constitutes one of the major challenges to be overcome before true high-throughput glycopeptide analysis can be achieved. The development of new glyco-related bioinformatics tools is thus of crucial importance to fulfill this goal. Here we present SweetNET: a data-oriented bioinformatics workflow for efficient analysis of hundreds of thousands of glycopeptide MS/MS-spectra. We have analyzed MS data sets from two separate glycopeptide enrichment protocols targeting sialylated glycopeptides and chondroitin sulfate linkage region glycopeptides, respectively. Molecular networking was performed to organize the glycopeptide MS/MS data based on spectral similarities. The combination of spectral clustering, oxonium ion intensity profiles, and precursor ion m/z shift distributions provided typical signatures for the initial assignment of different N-, O- and CS-glycopeptide classes and their respective glycoforms. These signatures were further used to guide database searches leading to the identification and validation of a large number of glycopeptide variants including novel deoxyhexose (fucose) modifications in the linkage region of chondroitin sulfate proteoglycans.

Assignment of coronavirus spike protein site-specific glycosylation using GlycReSoft

“…Therefore, specialized deconvolution algorithms are required for glycoproteomics data. SweetNET, a bioinformatics workflow for glycopeptide tandem mass spectral analysis [26] used the MS-DeconV algorithm for spectral deconvolution [27] and the MASCOT [28] for protein identification. The GPQuest glycopeptide spectral library search algorithm [29] used undisclosed isotope pattern fitting and spectral averaging methods for precursor mass calculation.…”

Data-independent oxonium ion profiling of multi-glycosylated biotherapeutics

“…These ions have been utilized in numerous advantageous approaches to qualitatively and quantitatively assess site-specific glycosylation. 16,17,23–28 However, oxonium ions have yet to be exploited to create agnostic profiles for high-throughput glycopeptide screening of multi-glycosylated therapeutics. Even with the substantial advancements made in glycoprotein analysis, it still remains difficult to unambiguously assign all potential sites of glycosylation and their associated glycans for complicated multi-glycosylated proteins.…”

Chemical Glycoproteomics

“…Likewise, bioinformatics platforms capable of extracting glycosites and structures from complex MS n data sets must keep up with new experimental technologies. (For more information, see the recent review by Hu et al 391 ) Commercial analysis tools such as Byonic (Protein Metrics) 392 and SimGlycan (Premier Biosoft); 393 several open source and academic analysis tools; 391,394 and database tools such as GlycoMod, Glyco-SuiteDB, 395 and UniCarbKB 396,397 are already having an impact in this regard. As well, big data studies correlating disease-associated mutations and genetic variations with PTMs are emerging and should help frame studies of the glycoproteome.…”