reading of a very instructive and easy-to-follow paper from
Pappireddi et al. [16].
18. One limitation of ESTs is that they rarely span the full mRNA
sequence (especially in the case of single-pass ESTs).
19. We refer in this section to relative quantification, i.e., determining the relative ratio of the amounts of every protein in
different samples, with the objective of determining fold
changes and differential expression of proteins across different
treatments or biological conditions, and not to absolute quantification, i.e., determining the concentration of a protein in a
sample, that can be done using other methods, such as including spikes of protein standards with known absolute
concentration.
20. The instrument variability and the bias in sample preparation
are different in every multiplex experiment, so it is recommended to design experiments using a number of samples
that fit within a single multiplex experiment whenever possible,
as all samples will be affected by the same amount of variability,
and relative quantification is rather straightforward, minimizing also the number of missing values (note that hyperplexing
strategies that allow up to 54 samples in the same run are being
developed [15]). Normalization methods presented here allow
comparison of results obtained from different multiplex experiments although it is important to bear in mind that technical
variability would still be higher when results are obtained from
multiple experiments compared to those obtained from a single
multiplex experiment.
21. Some proteomic software includes the possibility of internormalization among experiments using a reference channel.
However, we have noted that the default procedure is to apply
the same correction factor for all proteins of each experiment.
In our experience with pre-fractionated samples, technical
variability among experiments is different for each protein, so
a different correction factor should be applied for each protein.
If the chosen software does not allow for this possibility, it
would be necessary to download the quantitative data into a
spreadsheet and perform this normalization manually.
Acknowledgments
This work was supported by the Spanish “Ministerio de Ciencia
e Innovaci on” (codes BFU2011-22599, AGL2014-52062R and PID2019-107611RB-I00), Fondos Feder (ERDF,
European Commission), and Xunta de Galicia (“Grupos de Referencia Competitiva” ED431C 2020/05). The authors are very
Shotgun Proteomics in Non-model Organisms
99
Pappireddi et al. [16].
18. One limitation of ESTs is that they rarely span the full mRNA
sequence (especially in the case of single-pass ESTs).
19. We refer in this section to relative quantification, i.e., determining the relative ratio of the amounts of every protein in
different samples, with the objective of determining fold
changes and differential expression of proteins across different
treatments or biological conditions, and not to absolute quantification, i.e., determining the concentration of a protein in a
sample, that can be done using other methods, such as including spikes of protein standards with known absolute
concentration.
20. The instrument variability and the bias in sample preparation
are different in every multiplex experiment, so it is recommended to design experiments using a number of samples
that fit within a single multiplex experiment whenever possible,
as all samples will be affected by the same amount of variability,
and relative quantification is rather straightforward, minimizing also the number of missing values (note that hyperplexing
strategies that allow up to 54 samples in the same run are being
developed [15]). Normalization methods presented here allow
comparison of results obtained from different multiplex experiments although it is important to bear in mind that technical
variability would still be higher when results are obtained from
multiple experiments compared to those obtained from a single
multiplex experiment.
21. Some proteomic software includes the possibility of internormalization among experiments using a reference channel.
However, we have noted that the default procedure is to apply
the same correction factor for all proteins of each experiment.
In our experience with pre-fractionated samples, technical
variability among experiments is different for each protein, so
a different correction factor should be applied for each protein.
If the chosen software does not allow for this possibility, it
would be necessary to download the quantitative data into a
spreadsheet and perform this normalization manually.
Acknowledgments
This work was supported by the Spanish “Ministerio de Ciencia
e Innovaci on” (codes BFU2011-22599, AGL2014-52062R and PID2019-107611RB-I00), Fondos Feder (ERDF,
European Commission), and Xunta de Galicia (“Grupos de Referencia Competitiva” ED431C 2020/05). The authors are very
Shotgun Proteomics in Non-model Organisms
99