Algorithms for MS1-Based Quantitation

MS1-based quantitation is performed by direct integration of peptide precursor signal intensity from the MS1 spectra across retention time, based on the assumption that these signals have a linear relationship with abundance across a relatively wide dynamic range. Since ionisation efficiency varies between peptides, only relative abundance changes between biological samples are usually established. Whether each sample is run individually 'label-free', or two or three samples multiplexed within each run by a MS1-based labelling technique such as stable isotope labeling by amino acids in cell culture (SILAC), the informatics methods involved are broadly similar. In this chapter we present the key components of such pipelines, including the detection and quantitation of peptide features from the raw data, alignment of chromatographic variations between runs so that corresponding features can be matched, intensity normalisation to correct sample-loading differences and ionisation fluctuations, and methods to combine the peptide-level quantifications for the statistical analysis of differential protein expression across treatment groups. At each stage, the techniques have been designed for robustness against the systematic and random variation inherent in MS data, and errors during the preceding parts of the pipeline.