Naslov
Higher order structure and encoding of mass spectrometry data for high throughput human Fc IgG N -glycosylation analysis

Autori
Razdorov, Genadij ; Ugrina, Ivo ; Lauc, Gordan

Vrsta, podvrsta i kategorija rada
Sažeci sa skupova, sažetak, znanstveni

Izvornik
Glycoconjugate Journal / - : Springer, 2015, 240-240

Skup
GLYCO 23 XXIII International Symposium on Glycoconjugates

Mjesto i datum
Split, Hrvatska, 15.09.2015. - 20.09.2015

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Glyco(bio)informatics; glycomics; glycoproteomics; mass spectrometry

Sažetak
By continuous progress in mass spectrometry (MS) instrumentation, most importantly through increase in mass resolving power and scan speed, MS technology is generating more and more raw data. For high throughput MS, it is of crucial importance to properly structure and format generated data. Currently, MS experiments generate data encoded in binary proprietary formats, not accessible without producer’s library. Raw data is usually converted using ProteoWizard toolkit into XML based MS open formats, such as mzXML or mzML. These formats, including proprietary ones, are developed around individual MS experiment. When these tools are used in high-throughput setting, the main disadvantage is that data is spread into thousands of individual files, and is loosely structured (different number of scans per sample, different number of measurements in individual mass spectra, etc.). To enable high-throughput MS glycan analysis we developed an approach to restructure and reformat MS data into multidimensional (sample, scan, mass-over-charge and intensity) arrays encoded using hierarchical data format (HDF) library, designed for storage and organization of large amounts of numerical data. For this we used pyTables library that is build around HDF, but brings some additional advantages, such as BLOSC meta-compression, optimized for speed. Using this approach, we are able to efficiently analyze human Fc IgG N-glycosylation in relatively high throughput manner (cohorts of several thousand samples analyzed with a rate of roughly 100 samples per day).

Izvorni jezik
Engleski

Znanstvena područja
Biologija

POVEZANOST RADA