Biology Reference
In-Depth Information
and missing values in the dataset (see Note 5). Both approaches
have their unique strengths and limitations, but ultimately they
deliver essentially the same information.
Regardless of software approach, care should always be taken
to yield reproducible, high-resolution two-dimensional gel elec-
trophoretic separations, and this should include optimization steps
if necessary. In addition, this chapter assumes that the CyDye label-
ing, electrophoresis, and image acquisition procedures documented
elsewhere in this volume have been followed using proper experi-
mental design with a mixed-sample Cy2-labeled internal standard
to coordinate the appropriate number of biological replicates that
have been randomized with respect to dye labeling and gel position
to compensate for unanticipated dye or gel biases.
1. Inspect and import all gel images.
2. Detect feature boundaries and generate ratios on each gel inde-
pendently in the differential in-gel analysis (DIA) software
module, and then defi ne groups and set up experimental design
and matching between gels in the biological variation analysis
(BVA) software module (this can all be set up in batch mode if
desired). Normalized ratios are generated in BVA at this stage.
3. Assess matching on all images in BVA and manually adjust
matching in all images following prescribed methods associ-
ated with the software. Ratios are recalculated as matching is
adjusted within the dataset.
4. Ensure that the picking references are not included in the anal-
ysis by breaking matches for the picking references on at least
one gel in the matched set (no more than two should be neces-
sary, depending on how many mismatches you allow for in the
EDA analysis). This will avoid skewing of the PCA results due
to preferential fl uorescence of the reference markers.
5. Open the extended data analysis (EDA) software module and
begin a new project (the project must be closed in BVA for this
but can be reopened once the data have been imported into
the new EDA project).
6. Create in EDA a manual base set, with 100% of spot maps
where protein is present (no missing values, see Note 6) and
remove the unassigned spot maps.
7. Perform the following calculations:
(a) Differential Expression Analysis, ANOVA.
(b) Principal Component Analysis, select option to place spot
maps into the score plot and proteins into the loading plot.
(c) For hierarchical clustering (if desired), under Pattern
Analysis select spot maps vs. proteins for both Proteins
and Spot maps and exp groups.
3.3.1. Quick Guide:
DeCyder
Search WWH ::
Custom Search