Biology Reference
In-Depth Information
analyses. To enable confi dent protein identifi cations, ideally their
complete sequence information needs to be available or otherwise
“de novo” interpretation of mass spectrometric raw data is neces-
sary. Cutting-edge mass spectrometry enables high mass accuracy
analyses. This allows for de novo-sequencing of peptides and pro-
teins, respectively. The exploratory analysis can now be divided
into two possible strategies, the database dependent and indepen-
dent interpretation. The general goal is the identifi cation of as
many proteins as possible out of a complex protein mixture. With
the database-dependent approach only unambiguously identifi ed
sequences are considered. Consequently, peptides not identifi ed by
database-search most probably due to polymorphisms, low MS/
MS quality, or posttranslational modifi cations will be lost in this
strategy. However, these peptides can be very decisive with respect
to sample classifi cation [
3
]. In contrast, the database-independent
approach called MAPA is based upon an algorithm that groups all
peptide precursor ions with the same mass to charge ratio (
m
/
z
)
derived from high mass accuracy mass spectrometric raw data in a
data matrix without initial database search. At the same time, the
frequency of observed
m
/
z
-fragments according to the concentra-
tion of peptides is counted (spectral count) or ion intensity and
added to the data matrix. Furthermore, the retention time is con-
sidered for binning the correct
m
/
z
-precursors [
11
]. This algo-
rithm is called ProtMAX and can be downloaded from
http://
www.univie.ac.at/mosys/software.html.
The resulting data matrix
can be analyzed statistically—for instance using COVAIN [
5
]—to
rank peptide precursor ions according to their impact on sample
classifi cation. COVAIN can be downloaded from
http://www.
univie.ac.at/mosys/software.html
.
In Fig.
2
an example is shown
for different growth conditions—mixotrophic versus photoauto-
trophic—of the unicellular green alga
Chlamydomonas reinhardtii
.
Interesting candidates can then be identifi ed via database search or
de novo interpretation according to Hoehenwarter et al. [
3
]. This
allows also for the identifi cation of rarely detectable forms of pro-
tein modifi cations and polymorphisms (Fig.
2e
). For instance, the
combination of metal oxide affi nity chromatography (MOAC)
with MAPA enables the unbiased quantifi cation of phosphopro-
teins [
12
-
14
]. In summary, although the untargeted analysis only
allows for relative quantifi cation it enables a rapid overview of what
is changed at a systems response level. Moreover, this strategy also
delivers basic data for the targeted analysis—the MASS WESTERN.
4
MASS WESTERN and ProMEX
The best known method for a targeted analysis of specifi c
proteins out of a complex sample is based on antibodies (e.g.
Western Blot, Fig.
3
). Besides time-consuming and extensive
