Biology Reference
In-Depth Information
3.3 Spectral Count
Assessment of Protein
Groups
ReGrouper is a key tool that connects SEPro and PatternLab software
suite. Parameters for the generation of PatternLab's input (Sparse
Matrix and Index File) must be carefully designed in order to facilitate
further biological interpretation. Although it allows the generation of
quantitative data from multiple formats, we will focus on the analysis
of the spectral counts per protein groups in order to adhere to prin-
ciples of parsimony (i.e., the minimum number of protein sequences
that describes a set of peptides).
1. Copy and paste all SEPro files from each treatment (i.e., if
three biological replicates, three SEPro files) into different
directories. Name the directories with a number (we use 0 for
control and 1 for treatment, for instance).
2. In Regrouper's input panel, click “Add directory” and indicate
the location of the directory containing the SEPro files to be
processed. If desired, type the descriptions for each directory.
3. In “Processing parameters”, eliminate internal and decoy
sequences and click “Load”.
4. In “Spectral Counting Analysis”, select the following “Groups”
format: “Spectral Counts” and “Undetermined”. Then, press
“GO!”. With this, an Index file and a Sparse Matrix will be
created based on the distribution of spectral counts per pro-
tein groups that share the same peptide set independently of
the protein group type.
5. To verify the content and the quantitative values (i.e., spectral
counts) attributed for each protein group, select “generate
bird's eye view” and save the file.
PatternLab for Proteomics is a multi-tool proteomic statistical
analysis software [
29
]. Although other modules are available, we
will focus on the detection of the differentially expressed proteins
based on the T-Fold test, which integrates Student's
t
-test and fold
change values for the assessment of quantitative differences through
pairwise analysis (
see
Note 7
).
3.4 Detection of
Differentially
Expressed Proteins
1. Open PatternLab for proteomics, go to “Analyze” panel, and
select “T-Fold”.
2. Load the Sparse Matrix and the Index file generated previ-
ously by Regrouper.
3. Select normalization methods (
see
Note 8
).
4. Select the minimal number of replicates in which a protein was
detected to be considered in the analysis. This value depends
on the number of replicates used in the experiment. For con-
servative approaches type the number of replicates per treat-
ment. For instance, if five biological replicates were used, then
one should have five SEPro files and, for a conservative
approach, should select five as minimum number of replicas per
