Mixed effect modelling of proteomic mass spectrometry data by using Gaussian mixtures

WJ Browne, IL Dryden, Kelly Handley, S Mian, D Schadendorf

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

Statistical methodology for the analysis of proteomic mass spectrometry data is proposed using mixed effects models. Each high dimensional spectrum is represented by using a near orthogonal low dimensional representation with a basis of Gaussian mixture functions. Linear mixed effect models are proposed in the lower dimensional space. In particular, differences between groups are investigated by using fixed effect parameters, and individual variability of spectra is modelled by using random effects. A deterministic peak fitting algorithm provides estimates of the near orthogonal Gaussian basis. The mixed effects model is fitted by using restricted maximum likelihood, and a parallel fitting procedure is used for computational convenience. The methodology is applied to proteomic mass spectrometry data from serum samples from melanoma patients who were categorized as stage I or stage IV, and significant locations of peaks are identified.
Original languageEnglish
Pages (from-to)617-633
Number of pages17
JournalJournal of the Royal Statistical Society Series C (Applied Statistics)
Volume59
Publication statusPublished - 1 Jan 2010

Keywords

  • Multilevel modelling
  • High dimensional data
  • Gaussian mixture
  • Surface-enhanced laser desorption
  • Random effects
  • Proteins
  • Mass spectrometry
  • Biomarkers
  • ionization
  • Melanoma

Fingerprint

Dive into the research topics of 'Mixed effect modelling of proteomic mass spectrometry data by using Gaussian mixtures'. Together they form a unique fingerprint.

Cite this