- Abstract
- Technical Description
- Requirements
- How to run the Script
- Input Data
- Graphical User Interface
- Data Sets for Testing
- Output
- Troubleshooting
- Contributions
- Changelog
- References
- How to cite
- License
Middle-down proteomics aims to identify peptides in the range of 3-10 kDa. The tool analyzes the identified middle-down peptides and graphically displays various properties . The following analyses can be performed:
- Specificity analysis: Two-dimensional histogram of cleavage sites of the identified peptides. Determination of full-, semi- and un-specific peptides.
- Determination of the sequence coverage of all proteins.
- Analysis of N-terminal Methionine Excision.
- Visualization of identified peptides in respect to the full-length protein.
The entire script is based on the script "Potential Cleavage Analyzer" (PTK, 2021).
For each peptide, the tool searches the corresponding full length protein in the fasta database and the peptides are examined for their cleavage sites (Figure 1). If the peptide can be assigned to multiple proteins, the tool checks if the corresponding amino acids N- and C-terminal are the same for all assigned proteins. If this is not the case, the peptide is not considered further due to ambiguity. N-terminally non-truncated peptides (or with only start methionine cleaved) are used for start methionine cleavage analysis.
Figure 1: Determination of potential cleavage sites by analyzing the identified peptides. The tool searches for each peptide their corresponding full length protein in the fasta file and identifies the potential cleavage sites.
- Python 3.9.13 or higher
- Bio==1.5.3
- matplotlib==3.5.1
- pandas==1.4.2
- PyQt5==5.15.7
- pyteomics==4.5
- seaborn==0.11.2
Installation and use of Anaconda Distribution and its build-in command line prompt is highly recommended. In case you don't use Anaconda, make sure all required packages are installed upfront.
$ cd <PATH/TO/SOURCES>
$ python GUI.pyThe tool allows different input formats for the identified proteoforms:
- Proteome Discoverer database results. Exported "proteoform results" as txt-files. If the search was performed within Proteome Discoverer, use the "File->Export->To Text" function to export the proteoform results as txt-file.
Note: Currently only results obtained from ProSightPD searches are supported.
Fasta file, which was also used for database search. If database search was performed against a XML file, download the corresponding Fasta-file from UniProt.
Figure 2: Graphical User Interface.
- Fasta File. Opens file dialog to select a Fasta File, see Input Data for details.
- PD Result File. Opens file dialog to select a proteoform file, see Input Data for details.
- Select file format, see Input Data for details. Select PDProSight-Version.
- Select Protease used (currently only GluC is supported).
- Initialization: Reading of the input files and initialization of different analysis. This may take a while depending on the size of the input files. The progress of the calculations is displayed in the command line.
- Proteoform Information: Generates pie chart of identified proteoforms, with distinction between N-terminal truncated, C-terminal truncated, N- and C-terminal truncated and full length (=not truncated) proteoforms.
- Potential Cleavage: Generates heatmap showing the number of amino acids identified in X and X'-Positions of N- and C-terminal truncated subsequence proteoforms.
- Methionine Cleavage: Generates plots displaying the N-terminal methionine cleavage excision properties.
- Analze Fasta
- Export Data: Opens file dialog to select path were all results will be saved as txt-files.
- Specificity:
- Show Sequence Coverage:
- Visualize Sequence Coverage: Select the protein accession of interest and "Analyze Accession" generates plots to visualize the identified peptides in respect to the full length protein.
To test the script, you can download the datasets in the Datasets folder, which contains an human fasta-file and a Dataset Demo file containing peptide identifications exported from ProSightPD results. A detailed description of the expected output results after running the script is shown in Output.
Settings:
- Input fasta file "Dataset\20210709_Uniprot_Human_reviewed.fasta"
- Input proteoform file "DEMO_Human_GluC.txt"
- ProSightPD
- ProSightPD v4.2
As soon as 'Initalization' is pressed, the analysis starts. The progress of the calculations is displayed in the command line. If there are peptides in the input file that cannot be assigned to an accession number, they will be reported in the command line.
All charts can be exported in different file formats.
The tool identifies the X and X’ cleavage site of identified peptides, whereby X represents the N-terminal and X’ the C-terminal cleavage position, respectively.
Figure 3: Potential cleavage analysis plot. The heatmap shows the potential cleavage sites between two amino acids determined by the peptides, with the X site shown vertically and the X' site shown horizontally. The histograms show the total number of amino acids determined at the X' site (right) or at the X site (top).
Figure 4: Potential cleavage analysis command line output. The command line shows the loaded input files (fasta- and peptide txt-file), the number of considered peptides and the number of cleavage events.
The tool determines the specificity of the middle-down peptides, depending on the selected protease. For full specific peptides, both the N- and C-termini, for semi-specific peptides either the N- or the C-termini, and for un-specific none of the termini can be explained by the protease specificty.
Figure 5: Protease Specificity.
The tool calculates all theoretical possible cleavages of all proteins in the database (that is, it basically calculates the number of all dipeptide combinations in the database).
Figure 6: Theoretical possible cleavage events in the entire fasta file.
The tool analyzes the number of missed cleavages and displays them as a histogram. The missed cleavages are calculated based on the enzyme specificity. For example, for GluC digestion in ammonium bicarbonate buffer, all glutamate residues present in the peptide sequence are counted, except for the glutamate residue at the C-terminus.
Figure 7: Histogram of missed cleavages of the identified peptides.
The tool identifies the number of annotated (full-length), N-terminal truncated, C-terminal truncated, and N- and C-terminal truncated proteoforms.
Figure 8: Percentage of identified C-terminal, N-terminal, C- and N-terminal or non-truncated (full length) proteoforms displayed as a pie chart.
The tool visualizes the identified peptides of a given protein accession in terms of their localization within the protein sequence and their specificity (fully specific, semispecific N-terminal, semispecific C-terminal, non-specific).
Figure 9: Visualization of the identified peptides compared to the full-length protein. The x-axis shows the amino acid position of the full-length protein (as deposited in the fasta file). The identified peptide are plotted as bars along the y-axis. The color of each bar represents the specificity of the identified peptide: full specific (black), only N-terminal specific (orange), only C-terminal specific (red), unspecific (blue).
Figure 10: Visualization of the identified peptides compared to the full-length protein in the command line window. The first row shows the full-length protein (as deposited in the fasta file) and the following rows show the identified peptides. The columns show 1) the index of teh start amino acid of the identified peptide (compared to the full-length protein), 2) the index of teh end amino acid of the peptide, 3) the peptide sequence, and 4) the number of associated PSMs.
The tool assigns the identified peptides to proteins in the database (caution: this is not a sophisticated protein inference! The peptides can be assigned to multiple proteins), determines the number of assigned peptides, the sequence coverage and visualizes the position of the peptides in the protein sequence.
FIgure 11: Table Sequence Coverage.
PTK, Python script written, basic ideas
AT, Many ideas for data visualization
- v1.0.0 (February 2023) First release (based on "Potential Cleavage Analyzer")
The tool is available under the BSD license. See the LICENSE file for more info.








