Biology Reference
In-Depth Information
within the ProteoRed MIAPE repository; and (3) performing the
reverse operation, allowing users to export from MIAPE reports
into XML fi les for computational processing, data sharing, or pub-
lic database submission [
27
].
3
Data Collection and Annotation of Proteins
We would like to illustrate the two major proteomics workfl ows: (i)
protein extraction, electrophoresis (1-DE or 2-DE), cutting out
bands (by 1-DE) or spots (by 2-DE), and MS and (ii) protein
extraction and MS. The type-(i) approach is conventional, while
the type-(ii) approach has been developed in more recent years and
is also referred to as “shotgun proteomics” [
54
].
In proteomic analyses based on 2-DE (the type-(i) approach),
the refi ned and extracted proteins are separated in the range of
isoelectric point 3-10, in the fi rst dimension, and molecular weight
10-100 kDa, in the second dimension [
55
]. A spot is cut out from
the gel and the mass spectrum of the peptide is measured after
reductive alkylation, enzymatic digestion, and desalination. Based
on the obtained mass list and software searches against protein and
nucleic acid databases, we detect the amino acid sequences match-
ing the corresponding peptide mass spectrum data to identify the
protein. To improve sensitivity, as well as the conventional
Coomassie Brilliant Blue dye, fl uorescent staining is also used
extensively [
2
]. In this approach, we can use a gaseous phase pro-
tein sequencer instead of MS to determine the amino acid sequence.
The improvement of analysis precision and software develop-
ment for MS have made possible a proteomics analysis not reliant
on electrocataphoresis (the type-(ii) approach) [
56
,
57
]. In this
new approach, the refi ned or the extracted protein is digested with
enzymes and the mixture of peptides is separated by liquid chro-
matography (LC) based on their hydrophobicity. The peptides
eluted from the LC column are analyzed by MS, such as MS/MS,
which is directly connected with the LC. The fi rst MS separates
each peptide ion and the second MS decomposes the peptide into
fragments and determines the corresponding sequence from the
fragmentation pattern. Software identifi es the protein based on a
homology search against reference databases. There are two deri-
vations of this approach, labeling proteins before the analysis and
analysis without labeling.
At the annotation stage of identifying proteins, we should give
attention to the stability of the protein identifi ers. A recent report
revealed signifi cant differences in the identifi ers among the main
protein databases: the International Protein Index (IPI), the
UniProt Knowledgebase (UniProtKB), the National Center for
Biotechnological Information nr database (NCBI nr), and Ensembl
[
58
]. In the report, it was noted that some entries submitted to a
