288
- For quantitative analysis, collate spectra obtained for each sam-
ple group being tested into (high or low molecular weight)
data sets. Calibrate using the protein standard that was run on
each chip surface. - For experiments that involved comparison of profiles between
different treatments groups, normalize the spectra by total ion
count. This averages the intensity and adjusts the intensity
scale for all spots. Exclude grossly discrepant spectra from this
analysis. If normalization shows a more than threefold varia-
tion between spectra, reject outliers and repeat. - Analyze the normalized spectra using the software biomarker
wizard to generate peak sets across multiple spectra. The soft-
ware does a first pass in which well-defined peaks are identified
and a second pass in which low intensity peaks are detected
that correspond to these well-defined peaks (see Note 7). - Generate sample statistics for peaks identified using the bio-
marker wizard. The software calculates the mean and standard
deviation of peak intensity for each sample group and tests for
significant differences using the appropriate statistical method
(see Note 8). - For each sample group test performed, record a list of peaks
analyzed and p values obtained. All peaks with p < 0.05 (95%
confidence limit) are considered significant.
In order to isolate protein peaks for identification purposes, a num-
ber of separation and fractionation procedures are used. In the
scheme described in this section, samples with or without removal
of abundant proteins (see Note 9) are first fractionated by charge
and the fractions are subsequently separated by SDS-polyacrylamide
gel electrophoresis. At each step in the process, fractions can be
tested by SELDI MS using H50 and NP20 protein chips to con-
firm the optimal location of the peak of interest.
Protein identification is performed in the excised gel pieces.- Dilute the (serum/plasma) sample in equilibration buffer
(phosphate buffered saline, supplied with kit) and add to the
equilibrated Proteoprep column. - Elute unbound material by centrifugation using equilibration
buffer and bound material using elution buffer (0.1 M glycine-
HCl, tween 20, pH 2.5, supplied with kit). - Reconcentrate multiple depletions of plasma using
Ultrafree- MC microcentrifuge filters (supplied with kit). - Deplete the reconcentrated material further on the Proteoprep
column followed by a further reconcentration step.
Depleted plasma can be tested by SELDI-TOF MS to confirm
loss of known abundant protein peaks from the spectra.
3.3 Protein
Identification
3.3.1 Remove Abundant
Proteins
Peter Kelly