Method for generating negative samples based on relationship between peptides and source proteins

Abstract

This invention provides a method for generating negative samples based on the relationship between peptides and source proteins, comprising the following steps: First, the source protein is labeled. Then, based on the relationship between peptides and protein sequences, the global and regional characteristics of the protein sequence are defined. Next, regional and global characteristic indices are calculated. Then, using the regional characteristic indices, high-quality peptide regions are screened out, and negative samples are identified within the inner peptide regions. Finally, using the global characteristic indices, high-quality protein sequences are screened out, and negative samples are identified within the outer peptide regions.

Claims

1. A method for generating negative samples based on the relationship between peptides and source proteins, comprising: Indicate the protein source; By understanding the relationship between peptides and protein sequences, we can define the global and regional characteristics of protein sequences. Calculate regional characteristic indicators and global characteristic indicators; And using the aforementioned regional characteristic indicators, peptide regions are screened out, negative samples are found in the inner peptide regions, and using the aforementioned global characteristic indicators, protein sequences are screened out, negative samples are found in the outer peptide regions.

2. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 1, wherein the global characteristic indicators include cut-point saturation, peptide coverage, and peptide density.

3. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 2, wherein the peptide density, the peptide coverage, and the cleavage saturation are sorted from largest to smallest to quantify the confidence level of using regions within the protein sequence not covered by peptides as negative samples.

4. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 1, wherein the regional characteristic index includes defining high-frequency peptide regions and calculating and sorting peptide regions into quantitative indicators.

5. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 4, wherein defining a high-frequency peptide region includes selecting amino acid sites whose coverage count is greater than the median of the coverage count of the entire amino acid site as eligible amino acid sites, and consecutive amino acid sites form peptide map regions.

6. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 5, wherein the calculation and sorting of peptide interval quantification indicators are performed on the peptide map regions as units, and peptide density, peptide coverage, and cleavage saturation are calculated.

7. The method for generating negative samples based on the relationship between peptides and source proteins as described in claim 6, wherein the peptide density, the peptide coverage, and the cleavage saturation are sorted from largest to smallest to quantify the confidence level of using regions within the protein sequence not covered by peptides as negative samples.

Images