Analyze of Maldi-TOF Proteomic Spectra with usage of mixture of Gaussian distributions
Fragment książki (Rozdział w monografii)
| Status: | |
| Autorzy: | Plechawska-Wójcik Małgorzata, Polańska Joanna, Polański Andrzej, Pietrowska Monika, Tarnawski Rafał, Widlak Piotr, Stobiecki Maciej, Marczak Łukasz |
| Wersja dokumentu: | Drukowana |
| Arkusze wydawnicze: | 0,47 |
| Język: | angielski |
| Strony: | 113 - 120 |
| Efekt badań statutowych | NIE |
| Materiał konferencyjny: | NIE |
| Publikacja OA: | NIE |
| Abstrakty: | angielski |
| The article presents the method of processing mass spectrometry data, with detection of peaks by using the mixture of distributions. Spectra are firstly subjected to preprocessing, involving calibration, normalization, denoising and baseline correction. After that they are modeled with Gaussian distributions. Each distribution represents a single peak. Means of Gaussians describe m/z values of peaks while standard deviations represent their width. Third parameters are weights representing highs of individual peaks. The simulation presents usage of Expectation- Maximization algorithm to processing spectra with known m/z value of albumin and unknown m/z value of other compounds existing in the analyzed data sets. The developed algorithm was employed in identification of m/z values of proteins attached to the albumin with usage of the decomposition of Gaussian components. Searched m/z values were discovered with predetermined accuracy |
