Performs both Type I and Type II corrections in lipid class-related mass spectra. To this purpose, however, the dmz values (i.e. the absolute error in monoisotopic m/z estimation) need to be specified for each sum composition. Typically, such values are initially unknown. In this case, the user should employ "Lipic_It" rather than "Lipic" function. Indeed, "Lipic_It" follows a recursive algorithm leading to a reliable estimation of dmz values. After the last iteration, the Type I and Type II correction results are shown. Specifically, "Lipic_It" returns the values of the lipid Relative Abundances compared with both the "Most Abundant Lipid" and the "Lipid Class". The former indicates the percentage ratio between the signal intensity of a specific lipid sum composition and the highest intensity detected in the mass spectrum. The latter corresponds to the percentage ratio between the signal intensity of that sum composition and the overall intensity measured for the whole lipid class. If equal ionization yields are assessed, "Lipid Class" percent Relative Abundances can be considered equivalent to the mole fraction percentage of the sum compositions. In the "Lipic_It" output data frame, both "Most Abundant Lipid" and "Lipid Class" percent Relative Abundances are defined as "Adj" or "Un-Adj". Precisely, "Adj" (Adjusted) indicates that the relative abundances are computed after both Type I and Type II corrections. Conversely, Un-Adj (UnAdjusted) Relative Abundances are calculated using only Type I correction. The bar chart generated in the RStudio viewer pane at the end of "Lipic_It" calculations shows the interactive comparison between the "Adj" and "Un-Adj" values of "Lipid Class" percent Relative Abundances. This graphical output allows the user to perceive the extent of Type II correction in his experimental mass spectrum. Moreover, the reliability of the corrections performed by "Lipic_It" can be evaluated by requesting the comparison between the real mass spectrum and the simulated mass spectra obtained after either Type I or both Type I and Type II corrections. Subsequently, a new plot is generated in the RStudio viewer pane. Here, these spectra are named "Uncorrected Simulated Spectrum" (Type I correction) and "Corrected Simulated Spectrum" (Type I and Type II correction). The user can also choose which spectra need to be removed from spectral superimposition by clicking on the corresponding line in the legend panel.
Lipic_It(data, z, Res, t = 0.001, span = 25, ppm = 10, xseq = 0.001)
| data | A data frame reporting the recognized sum compositions in the first column. Conversely, the corresponding chemical formulas, the measured m/z, the dmz, and the absolute intensity values need to be listed in the second, third, fourth and fifth columns respectively. N.B. All the elements in the data frame must be sorted following the ascending order of the measured m/z value. Furthermore, the "dmz" column must be filled with 0. Please, check the "CLmix" data frames included in the LIPIC package as an example of data organization of the "Lipic_It" input data frame. |
|---|---|
| z | Charge of the lipid ions. The charge sign needs to be specified. |
| Res | Operating resolution of the mass analyzer. |
| t | Probability below which isotope peaks can be omitted. Refer to the "isopattern" function in the "enviPat" package for details. |
| span | The "span" parameter value required by the "pick.peaks" function. This function is included in the "LIPIC" package, but belongs to the "ChemometricsWithR" package. |
| ppm | Accuracy threshold (in ppm) required by the "dmzFind" function. |
| xseq | Increment in the sequence of the m/z values used for the calculation of the simulated mass spectrum employed in dmz calculation. "xseq" is set to 0.001 as default. Higher values reduce the computation time, but can lower the accuracy by which the dmz values are retrieved by the "dmzFind" function. |
A data frame which is an extension of the input data frame. The first five columns are identical to those of the input data frame, except the dmz column (i.e. the fourth column), which contains the dmz estimated values for each sum composition. The accuracy of "Lipic_It" correction is strictly dependent on the reliability of dmz values. These are recursively calculated until convergence and the overall content of the output data frame reflects the results of the last iteration. Precisely, the sixth column shows the intensity values after Type II correction. The seventh column shows the intensity values after both TypeII and Type I corrections. As indicated by the corresponding headers, the remaining columns show the "Adj" and "Un-Adj" "Lipid Class" percent Relative Abundance and the "Adj" and "Un-Adj" "Most Abundant Lipid" percent Relative Abundance (see the above description section for details).
Andrea Castellaneta, Ilario Losito, Davide Coniglio, Beniamino Leoni, Pietro Santamaria, Maria A. Di Noia, Luigi Palmieri, Cosima D. Calvano, Tommaso R.I. Cataldi: "LIPIC: an automated workflow to account for isotopologues-related interferences in electrospray ionization high resolution mass spectra of phospholipids", Journal of the American Society for Mass Spectrometry, 2021. DOI: https://doi.org/10.1021/jasms.1c00008.
Martin Loos, Christian Gerber, Francesco Corona, Juliane Hollender, Heinz Singer: "Accelerated Isotope Fine Structure Calculation Using Pruned Transition Trees" Anal. Chem. 87, 5738–5744 (2015). DOI: https://doi.org/10.1021/acs.analchem.5b00941.
Ron Wehrens: Chemometrics with R. Springer Verlag, Berlin Hei-delberg (2020). DOI: https://doi.org/10.1007/978-3-642-17841-2.
Carson Sievert: "Interactive Web-Based Data Visualization with R, plotly, and shiny" Chapman and Hall/CRC, 2020, https://plotly-r.com.