Q&As About Proteomics (Q16-Q20)

Q16. What bioinformatics analyses do the proteome results usually need?

At present, the analyses in general literature mainly include Venn diagram analysis, heatmap analysis, volcano plot analysis, hierarchical cluster Analysis, protein function analysis (GO classification), protein metabolic pathway enrichment analysis (KEGG-based pathway), differential protein interaction network construction, transcription factor prediction, etc.

Other relatively high-level analyses include multiple sets of data association analysis, multi-pathway extension analysis, phosphoprotein kinase predictive analysis, interaction networks, and biocontrol model construction analysis.

Q17. What are the differences between iTRAQ and other proteomics techniques?

Two-dimensional electrophoresis is the earliest and most classic proteomics technique. It is suitable for most samples, but strong acid or strong basic protein, and proteins whose molecular weight are too large or too small. In addition, as each sample needs to run alone, it is difficult to avoid the influence of operational error on the results, which may lead to inaccurate quantification. However, the service price of two-dimensional electrophoresis is relatively cheap and can be used for preliminary screening of sample differences.

DIGE has been improved on the basis of two-dimensional electrophoresis, and fluorescent markers and internal standards have been introduced to make the quantification more accurate. However, this technique is also based on the isoelectric point and molecular weight of the protein, so it doesn’t have a good separation effect on the strong acid or strong basic protein, and proteins whose molecular weight is too large or too small.

iTRAQ, TMT, label-free, and Silac all use LC/MS technology to find differential proteins. Among them, the principle and detection method of iTRAQ and TMT are basically the same, except that different markers are used. TMT has 10 kinds of markers and can analyze up to 10 different samples, but as the mass of the markers is very similar, ultra-high resolution mass spectrometry must be used to satisfy the detection, which will inevitably lead to more signal interference and may affect the accuracy of quantification. Label-free is a non-labeled proteomics technique. Because there is no label, its quantification depends on the stability of the operation and the mass spectrometer. The number of proteins identified by this technique is less than that of iTRAQ technology, and the accuracy of quantification is poor. The advantage is that the service price is low. SILAC is a proteomics method based on stable isotope-labeled cell culture technology. It uses different isotopic mediums to culture different groups of cells, achieving the purpose of in vivo labeling. Its quantitative accuracy is better than other proteomics technology. However, due to the lack of a suitable isotopic medium, this method can only be used for mammalian cell lines that can be passaged.

From the aspects of quantitative accuracy and applicability, iTRAQ technology has become the most widely used proteomics technology in recent years.

Q18. Can different types of samples be used for iTRAQ at a time?

No. If the sample is from a different species, the database cannot be selected at the time of retrieval, and the data cannot be analyzed. Even for samples of the same species, their protein types and abundances may vary greatly (such as plant roots and leaves). These enzymatic peptides are mixed on the machine and their data interfere with each other, resulting in inaccurate protein quantification and characterization.

Q19. When the data of the iTRAQ experiment is searched by different databases, why are the ratios between the groups different?

The ratio between the different markers in iTRAQ technique is related to the retrieval database, the noise of the mass spectrometer electrical signals, and the randomness of the acquisition of the mass spectrum. Therefore, when using different database searches, the quantitative results are slightly different due to the difference in the matching peptides, while the overall trend should be consistent.

Q20. Can species that are not sequenced be a proteome?

As a large number of species have been sequenced and the corresponding genes are predicted, unsequenced species can be used for proteome research through the genetic sequences of closely related species as a database. Therefore, in general, unsequenced species can be considered as a database by considering the genomic sequence of a closely related species, and can be used as a proteome.