Summary:
- Definition of XIC AUC and XIC Ratio% Byos PTM values/report fields
- Walkthrough of the three Blgc_PTM (Default).rptc percent modification calculations
- Review of some Byos reporting fundamentals
Introduction:
The default Byos PTM report, called Blgc_PTM (Default).rptc, includes three report tabs for automatic calculation of relative percent Post-Translation Modification:
1. %Mod by ModName across Samples
2. Avg %Mod per AA by Sample
3. % Mod (Multi Mods per Peptide)
While the default approaches above are suitable for many users, Byos reports can be customized to calculate PTM percent modification in a variety of ways. Customizing PTM percent modification calculations is beyond the scope of this article.
This article reviews the default PTM report tabs and explains how percent modification is calculated for each tab using a simplified example project including:
- three modifications: Oxidation at M135, Dioxidation W196 and Deamidation at N198 (N>D and N>isoD),
- two peptide sequences: DTLMISR and VVSVLTVLHQDWLNGK,
- and four forced degradation treatment time point samples: Day0, Day1, Day3, and Day7.
XIC Ratio% and XIC AUC:
XIC AUC is the area under the XIC curve. XIC AUC is measured for each charge state of each peptide identified. The XIC is plotted by measuring the abundance of a single isotopic peak, typically the most abundant isotope, over time. Figure 1 below shows the monoisotopic (iso_0) XIC plot of charge state +1 (z_1) for the unmodified peptide DTLMISR. The measured XIC AUC is 6.137e+07.
Figure 1: Byos Project Investigation XIC and Isotope plots for wildtype DTLMISR (z=1).
XIC Ratio% is the percent XIC AUC of a modified peptide relative to its corresponding wildtype XIC AUC of the same charge state and is calculated using the equation below.
Figure 2 below shows XIC AUC and XIC Ratio% values for oxidized DTLMISR and wildtype DTLMISR, both observed at charge states +1 and +2. XIC AUC (in red boxes) is measured for all four peptides signals. XIC Ratio% is empty for wildtype peptide signals and modified peptide signals lacking a corresponding wildtype identification of the same charge state. XIC Ratio% (in green box) is calculated for each oxidized DTLMISR charge state using the equation above. Using oxidation of DTLMISR (z=1) as an example:
(6.22e+06) / (6.22e+06 + 6.14e+07) = 9.2%.
Figure 2: A few rows of the Byos Project Investigation Peptides Table. The table is filtered to only show a single sample. Dropdowns are expanded to show all four peptide signals corresponding to wildtype (z=1 and z=2) and oxidized (z=1 and z=2) forms of the peptide. Boxes are added to highlight XIC AUC (red) and XIC Ratio% (green).
%Mod by ModName across Samples
The %Mod by ModName across Samples report tab calculates unweighted averages of XIC Ratio% using all peptides and charge states corresponding to each modification. Figure 3 shows the report tab with a row for each modification and column for each sample.
Figure 3: The %Mod by ModName across Samples report tab. The right-most columns show the report tab output cells, or percent modification values calculated for the four modifications and four samples included for this article (red box).
To understand how the report tab is constructed, Figure 4 shows the report tab configuration details, viewed by clicking the Show configuration button, along with colored boxes highlighting key sections of the report tab configuration.
Figure 4: Configuration details of the %Mod by ModName across Samples report tab.
- Purple: The report tab type is set to Table Barchart.
- Gold: The pivot table column and row fields are selected to separate the columns by sample and the rows by modification. Note the absence of Sequence or Charge fields. This means the XIC Ratio% of all sequences and charge states corresponding to a single modification for a single sample will be aggregated together. A filter is applied to Mod. Names to remove all unmodified/wildtype values from the report tab.
- Red: The aggregator is set to Average and XIC Ratio%. This means multiple XIC Ratio% values aggregated to the same output cell will be averaged.
- Green: A single M135 oxidation percent modification value calculated the Day0 FD treatment sample. Figure 5 breakdowns all aggregated XIC Ratio% values averaged to calculate the percent modification value of 5.63% (discussed further below).
Adding Sequence (unformatted) and z fields to the pivot table as columns separates and reveals all individual XIC Raito% values averaged to calculate percent modification in the %Mod by ModName across Samples report tab. Figure 5 shows XIC Ratio% of DTLMISR z=1 (4.73%) and z=2 (6.53%), which are averaged to calculate percent modification of oxidation at M135 for the Day0 FD treatment sample (see the green box in Figure 4) as shown in calculation below.
5.63% = (4.73% + 6.53%) / 2
NOTE: Absolute abundance of z=1 and z=2 charge states are not accounted for when averaging the XIC Ratio% values, meaning the %Mod by ModName across Samples report tab calculates an unweighted average of all charge states to determine percent modification.
Figure 5: Sequence (unformatted) and z are added to the pivot table, separating the two XIC Ratio% values that are averaged to calculate percent modification for oxidation of M135, FD Day0 FD treatment sample.
Avg %Mod per AA by Sample
The Avg %Mod per AA by Sample report tab calculates percent modification at each Modified Amino Acid (Mod. AAs) by averaging XIC Ratio% across all peptides, charges, and modifications (Figure 6). The report tab is useful for high level comparison of overall percent modification observed between samples.
Figure 6: Avg %Mod per AA by Sample bar chart for the simplified example used in this article. Red, blue, and orange columns show the average XIC Ratio% of all modifications observed at amino acids M, N, and W, calculated and plotted for each sample (Day0, Day1, Day3, and Day7).
To understand how the report tab is constructed, Figure 7 shows the report tab configuration details, viewed by clicking the Show configuration button, along with colored boxes highlighting key sections of the report tab configuration.
Figure 7: Configuration details of the Avg %Mod per AA by Samples report tab.
- Purple: The report tab type is set to Bar Chart.
- Gold: The pivot table column and row fields are selected to separate the columns by sample and the rows by Mod. AAs. Note the absence of fields to separate values by anything other than the modified amino acid. This means the XIC Ratio% of all modifications, modification sites, proteins, sequences, and charge states corresponding to a single modified amino acid for a single sample will be aggregated together. A filter is applied to Mod. AAs to remove all unmodified values from the report tab.
- Red: The aggregator is set to Average and XIC Ratio%. This means multiple XIC Ratio% values aggregated to the same output cell will be averaged.
- Green: A single N percent modification value calculated the Day0 FD treatment sample. In Figure 8 we breakdown all XIC Ratio% values averaged to calculate the percent modification value of 0.428%.
Converting the report tab from a Bar Chart to a Table and adding the Mod. Names, Protein Name, Var. Pos. Protein, Sequence (unformatted), Labels, and z fields as columns shows the individual XIC Ratio% values averaged to calculate percent modification at each modified amino acid (Figure 8). For the example used in this article, percent modification at Asparagine is determined by averaging the four XIC Ratio% values included in the example report: Deamidation N>D z=2 (0.510%), Deamidation N>D z=3 (0.282%), Deamidation N>isoD z=2 (0.531%), and Deamidation N>isoD z=3 (0.390%).
0.428% = (0.510% + 0.282% + 0.531% + 0.390%) / 4.
Figure 8: Mod. Names, Protein Name, Var. Pos. Protein, Sequence (unformatted), Labels, and z are added to the pivot table, separating the four XIC Ratio% values that are averaged to calculate percent modification of the modified amino acid N for the Day0 FD treatment sample.
% Mod (Multi Mods per Peptide)
NOTE: {XIC area summed isoX normalized} is the same as XIC AUC for all wildtype peptides and most modified peptides, and only differs for some deamidated peptides. Using {XIC area summed isoX normalized} for the % Mod (Multi Mods per Peptide) report tab ensures XIC AUC values are normalized to the same isotope peak. For more details please see the knowledgebase articles: Asparagine Deamidation Quantified and Byos PTM Quantitation Isotope Selection and {XIC area summed isoX normalized}.
For the purposes of this article and most percent modification calculations in Byos PTM, {XIC area summed isoX normalized} can be considered the same value as XIC AUC.
The %Mod (Multi Mods per Peptide) report tab groups modifications by peptide sequence and reports percent {XIC area summed isoX normalized} of each modification relative to the sum of {XIC area summed isoX normalized} for all modifications and wildtype peptide signals with the same the peptide sequence. The calculated percent modification for each modification is a weighted average of all charge states identified. Unlike XIC Ratio%, each percent modification value calculated in the % Mod (Multi Mods per Peptide) report tab is determined relative to abundance of the wildtype and other modifications identified for the peptide sequence. Figure 9 shows the report tab with a row for each modification and column for each sample.
Figure 9: The % Mod (Multi Mods per Peptide) report tab for the simplified example used in this article. The right-most columns show the report tab output cells, or percent modification values calculated for the four modifications and four FD treatment samples included for this article (red box).
To understand how the report tab is constructed, Figure 10 shows the report tab configuration details, viewed by clicking the Show configuration button, along with colored boxes highlighting key sections of the report tab configuration.
Figure 10: Configuration details of the % Mod (Multi Mods per Peptide) report tab.
- Purple: The report tab type is set to Heatmap with Color By set to {XIC area summed isoX normalized}.
- Gold: The pivot table column and row fields are selected to separate the columns by sample and the rows first by peptide sequence and then by modification. Note the absence of the z field. This means the {XIC area summed isoX normalized} of all charge states corresponding to a specific peptide sequence and modification for a single sample will be aggregated together. A filter is applied to Mod. Names to hide all unmodified/wildtype values from the report tab. Importantly, the filter only hides the wildtype rows, while keeping the wildtype {XIC area summed isoX normalized} values as part of the normalization calculation.
- Red: The aggregator is set to Normalize Column, {XIC area summed isoX normalized}, Sum, and Level 1. Selecting Sum and {XIC area summed isoX normalized} means multiple {XIC area summed isoX normalized} values aggregated to the same output cell will be summed together. Selecting Normalize Column Level 1 means normalization is performed for each unique value in row 1 of the report table. (Level 1 tells the report to normalize based on the first report table column, Level 2 would normalize based on the second report table column, etc.). The normalization calculation is explained in detail with an example in the section below.
- Green: A single W196 Dioxidation percent modification value calculated for the Day0 FD treatment sample. Figure 11 breakdowns each individual {XIC area summed isoX normalized} value used to calculate the percent modification value of 0.308% (discussed further below).
Adding the z and {XIC area summed isoX normalized} fields as pivot table columns and showing the wildtype rows by removing the hide filter on Mod. Names separates and displays all individual {XIC area summed isoX normalized} values used to calculate percent modification in the % Mod (Multi Mods per Peptide) report tab (Figure 11).
Figure 11: {XIC area summed isoX normalized} and z are added to the pivot table, separating all {XIC area summed isoX normalized} values used to calculate percent modification values. The green box highlights the peptide signals for Dioxidation of W196 identified in the Day0 FD treatment sample. The red box highlights all values grouped together for the Level 1 Column normalization.
With a Level 1 Column Normalization for {XIC area summed isoX normalized}, each output cell will show {XIC area summed isoX normalized} for the row divided by the sum of all {XIC area summed isoX normalized} values corresponding the same Column 1, Level 1, or Sequence (Unformatted) value. The sum of all VVSVLTVLHQDQLNGK {XIC area summed isoX normalized} values is:
207,938,399 + 50,311,804 + 1,064,883 + 142,133 + 1,110,918 + 196,830 + 747,231 + 59,642 = 261,571,840
Figure 11 shows {XIC area summed isoX normalized} of VVSVLTVLHQDQLNGK Dioxidation z=2 (747,231) and z=3 (59,642). The normalized output for the two charge states of VVSVLTVLHQDQLNGK Dioxidation are shown below. The report tab format is set to Percentage, meaning all output values are multiplied by 100.
100 * (747,231) / (261,571840) = 0.286%
100 * (59,642) / (261,571,840) = 0.023%
Unlike Figure 11, the % Mod (Multi Mod per Peptide) report tab does not include the z and {XIC area summed isoX normalized} columns. Without these columns and because the report tab aggregator is set to Sum, the {XIC area summed isoX normalized} individual values of charge states 2 and 3 for VVSVLTVLHQDQLNGK Dioxidation are summed together.
747,231 + 59,642 = 806,873
The Level 1 Column normalization calculates percent modification for VVSVLTVLHQDQLNGK Dioxidation for the Day0 FD Treatment sample:
100 * (806,873) / (261,571,840) = 0.308%
We can see 0.308% matches the reported percent modification value highlighted with the green box in Figure 10.