24 A Practical Guide to Cancer Systems Biology
derived from the digestion of the proteins.^4 Shotgun proteomics, also called
peptide-centric or “bottom-up” proteomics, is widely used for proteomic
investigation. The aim of proteomics is to accurately characterize as many
proteins as possible, and the ultimate goal is the explicit information of
the complex mixture containing at least 10,000 different proteins in a cell
population.^5
The generic workflow of shotgun proteomics starts from the digestion
of proteins into peptides by sequence-specific enzymes such as trypsin. The
peptides are subsequently separated by on-line Ultra Performance Liquid
Chromatography (UPLC) and converted into gas phase ions by electrospray.^6
The peptide ions are scanned by mass spectrometer to generate mass
spectra, and selected peptides from the mass spectra are subjected to
collision-induced dissociation (CID) for fragmentation. The fragment ions
are measured by the second mass spectrometer, and the fragment ion spectra
generated are assigned to their corresponding peptide by sequence database
search algorithm. Finally, protein inference is performed by assigning peptide
sequences to proteins.
Quantitative proteomics is the identification of proteins and mea-
surement of protein abundance in different samples. The profiling from
quantitative proteomics enables us to compare the proteomes of samples
in different biological status, such as altered expression of proteins in
response to a specific stimulus. The proteins showing changes in abundance
are possibly associated with the biological processes that determine the
different status of the samples. Two major methods are available for
performing quantitative analysis of MS-based shotgun proteomics, the label-
free quantification and stable isotope-based quantification.^7 The label-free
quantification can compare across many samples in one experiment, but the
accuracy is lower and not sensitive enough to small changes in abundance.
The isotope-based quantification has the advantage of accuracy, but it is
only used for comparison of up to eight samples at a time. The data of
quantitative proteomics have two forms, the absolute amount of each protein
in a sample or the relative change in protein amount among different samples.
The absolute quantification or AQUA^8 is achieved by comparing the signal
of the peptides from the protein of interest to a spiked in or isotope-labeled
peptide. The relative quantification is more well established compared to the
absolute one that required time-consuming processes of developing reference
materials and assay condition for proteins.^9
One of the label-free quantification is achieved by spectrum counts,
inferring protein abundance by the number of times a peptide is observed
and the number of distinct peptides observed from a given protein. Another