|
Abstract:
|
Expression proteomic analyses that use profiles generated by Matrix Assisted Laser Desorption Ionization, Time-of-Flight mass spectrometers (MALDI-TOF MS) require three components: quality control, feature extraction, and inference. We simultaneously address the components by assuming an over-complete functional data analysis approach that capitalizes on an underlying levy process and uses information across multiple spectra. Specifically, we assert that one, m/z, and resolution-dependent stochastic gamma process influences every within-population, multi-modal spectrum; yet, random, biological, or measurement error force spectra to deviate from the process parameters. Under this assertion, a Bayesian hierarchical approach naturally models data quality control variables and peak parameters while leading to posterior predictions of experimental-group status.
|
