Algorithms for automated identification and quantification of glycans and glycopeptides.

摘要

Glycosylation is one of the most common post-translational modifications in which glycans are covalently linked to the sidechain of Asn (i.e., the N-linked glycosylations) or Ser/Thr (i.e., the O-linked glycosylations) residues. The glycosylation alterations are known to be associated with various biological processes and human diseases, to understand which it is essential to study this biological phenomenon. Liquid Chromatography coupled with Mass Spectrometry (LC-MS) is a rapid and sensitive analytical method that has already been deployed in biological research for several decades. Several mature protocols have been developed for glycomics and glycoproteomics; however, there is a lack of suitable bioinformatics tools for the automated analysis of these data. The aim of this study is to develop automated algorithms to assist glycomics and glycoproteomics research. The first part employs high confidence glycan profiling using LC-MS; the algorithm utilizes the characteristics of LC data to increase the number of true identifications. Both m/zs and elution profiles are considered while assigning glycan, and the elution time prediction model is employed in order to distinguish among glycans that have close m/z but different compositions. Because the algorithm can report glycan profiling results with high confidence, it is the foundation of glycan quantitation. The second part of this study focuses on glycan relative quantitation by using either a labeling or a label-free technique. The label-free protocol can quantify as many samples as a user needs, and the algorithm can automatically adjust the quantified ratio for imbalanced data. The last part focuses on glycan sequencing, which depicts the topology of N-linked glycan but without linkage information. The iterative algorithm annotates the spectrum by facilitating fragments resulting from collision-induced dissociation (CID), which comprises the majority of breakages of glycosidic bonds, and coupling with extra information from high-energy C-trap dissociation (HCD). The results from glycan profiling and quantification can be further used in glycoproteomics studies to narrow down or target the most important glycopeptides. In conclusion, the methods reported here provide a bottom-up analytic informatics solution for glycomics and glycoproteomics studies in complex samples.