What are the technical characteristics of label-free quantification?
- No labeling required; easy to operate
- Not limited by the number of samples to be compared
- Requires high stability and reproducibility of experimental operation
- Slightly less accurate than labeled quantification
- Requires at least three technical or biological replicates
Why can label-free compare different types of samples?
Label-free is a non-isotope labeling technique. There is no isotopic label within the sample, and each sample is individually digested and then subjected to chromatographic separation and mass spectrometric analysis independently in turn. The relative quantification of the proteins corresponding to the peptides is then performed by comparing the signal intensities of the corresponding peptides in the different samples. Thus, any sample type can be compared regardless of the same protein or gene.
What is the role of HPLC grading? Is the flux higher with more components?
HPLC grading reduces the complexity of the proteins in each fraction and increases the amount of each protein in each fraction, which facilitates mass spectrometry identification. We need to grade samples according to their complexity. Too few grades will not reduce the complexity of the sample, while too many grades will reduce the amount of each protein in each fraction.
Why is label-free not as accurate as iTRAQ?
When the sample is too complex, there are too many similar m/z, resulting in ineffective differentiation of the primary mass spectra, which eventually leads to abnormal peak area extraction and produces inaccurate quantification.
How should I analyze the data from three replicate mass spectrometry experiments that have zero quantitative values?
The characteristic of mass spectrometry is that if the same sample is subjected to multiple replicate mass spectrometry analyses, the results identified by mass spectrometry will not be identical each time (about 70% for Overlap). Whenever a protein is identified in one of the mass spectrometry results, it means that the protein is present in the sample. However, for the purpose of subsequent statistical analysis, the general criterion is to select those identified in at least two of the three mass spectrometry replicate experiments (i.e., data with quantitative values and non-zero quantitative values) and take the average. If a value is available in only one experiment, the data is not used for quantitative analysis.
When a protein in a sample group has no value (or shows "0") in three quantitative results, and the protein has a non-zero quantitative value in another group at least twice, the difference can be considered as "with or without" expression difference. However, the conclusion of "with or without" differences in mass spectrometry results must be made with caution and requires further verification by other experimental methods.
Why are some of the quantitative data "0"? What happens when there are qualitative but no quantitative results?
For label-free projects, we use the iBAQ or LFQ algorithms of MaxQuant software for quantitative analysis. Both algorithms are based on MS1 and calculate the integration of each peptide signal on LCMS chromatography. iBAQ is relatively newer and more widely used, so in general we provide quantitative results from iBAQ algorithm.
Reason 1: There are two types of peptide attributes, those that can be used for protein characterization and those that are unique for protein quantification, and the number of qualitative peptides ≥ the number of quantitative peptides. If there is no unique peptide that can be used for quantification of the identified protein, the protein has no quantitative value. In this case, the intensity of all samples is shown as 0.
Reason 2: If the signal value of the peptide used for protein quantification is too low in some samples, it leads to a situation where the peak area integral quantification has no value, and thus the intensity in that sample is shown as 0.
What are the criteria for determining the confidence of protein identification?
Our label-free project generally uses MaxQuant software for protein characterization and quantitative analysis. FDR is calculated by searching the Targetdatabase and the Decoy library (the Decoy library is automatically created by MaxQuant software), and the number of matching profiles is obtained. FDR≤0.01 is the accepted standard for screening histological data. The data in our report have been screened by the FDR≤0.01 criterion, so they are reliable data.
What are the basic principles for screening differential proteins?
There are no uniform criteria for screening differential proteins for quantitative protein information obtained from lable-free experiments. If the number of differential proteins obtained by screening with this criterion is high, the screening criteria can be increased according to the actual situation or needs, such as fold change > 2 or even > 4 or p value < 0.01, etc. If the experimental design is a technical replicate (mass spectrometry replicate assay), it is recommended to increase the difference filtering criteria to at least twofold.
How can I analyze the reported proteins with only protein registration numbers but no protein names or gene names?
The protein database used for mass spectrometry is usually a public protein database such as Uniprot, NCBI or a self-built library of amino acid sequences provided by the customer (Fasta standard format is required). Therefore, only protein information is presented in the protein identification table, not the corresponding gene information, and the protein identification table output by MaxQuant only provides the protein registration number but not the protein name or gene name. If you need the relevant protein or gene information, you can search for it on the corresponding database website according to the protein ID provided under the "Fasta headers" entry in the results.
