Functional genomics techniques (transcriptomics, proteomics and metabonomics) are becoming increasingly important in the life science. The aim of these techniques is to gain new insights and a better understanding of the biological functioning of a cell or organism.
Metabolic profiling of humans, animals and plants is proving increasingly important for understanding disease and monitoring the effects of drug treatment, nutritional regimes and toxicity. Metabolomics is the study of plant metabolites and is useful for varietal studies and the effect of growing conditions or genetic modification. Metabonomics is the study of animal or human metabolism on body fluids such as urine. The identification and quantification of changes in proteins related to disease or disease-modifying agents is part of Proteomics.
In all of the above research fields, Mass Spectrometry (LC/MS) plays an increasingly important role, due to its high sensitivity, rapid analysis and ease of identification, using exact mass.
The MsXelerator® software offers a large collection of tools, algorithms and visualization techniques for what in general could be called: Comparative Analysis.
MsXelerator offers all of the operations needed for processing LC/MS datasets:
- Pre-processing: binning, smoothing, baseline correction, de-spiking
- Peak detection and quantification
- De-isotoping, charge deconvolution and peak matching
- Alignment of mass chromatograms
- Normalization
- Multivariate Analysis, PCA/Clustering
- Classification algorithms and Biomarker Surface Maps
- Biomarker Discovery based on non-peak-detection methods: operate on raw mass chromatograms or mass spectra (local and full)
- Classification and Biomarker Discovery using peak-detection/matching algorithm
- Validation and Visualization
 |
Figure 1: LC/MS Data Processing –Different Analytical Data Modes can be used to detect unique features - A: Use Peak Detection methods that directly operate on the raw data, B: Local Screening using average mass chromatograms created from small m/z ranges, and C: Local screening using average mass spectra taken over small time regions (TCM/BCM) |
| |
Different Data Directions: |
|
Given a number of samples from different groups or classes, the MS Compare module can handle different data directions to search for unique features or peaks across the full 2D dataset:
- Raw data analysis: every data point will be used in the analysis and no assumptions are made whether the feature is in fact a chromatographic peak or not.
- Peak Picking: MPeaks will be used to detect significant chromatographic peaks in each sample. Using all samples, peaks are then re-ordered and matched. A new data matrix will be created holding all intensities for all peaks and samples.
- TCM/BCM Mass Spectra processing. Average Mass Spectra are calculated from small cross-sections (typically using 25-50 time windows). Mass spectra processing can be extremely useful when time shifts between samples are very complex and difficult to correct using the available alignment algorithms.
|
|
|
MS Compare:
Comparative analysis on series of samples is typically done using the MS Compare module. MS Compare offers a large set of visualization tools and all of the above mentioned algorithms: pre-processing, alignment, multivariate analysis, detection of unique features, biomarker discovery, reporting, etc. |
|
| |
Figure 2: Data Exploration in MsCompare. Bottom – overlay of TIC’s from 14 samples, Top – Extracted Ion Currents for all samples (Click and Identify). From the EIC plot it is seen that large time shift are present. |
Data Alignment:
The MS Compare module contains 4 alignment algorithms (offset shifting, cross correlation, peak based warping and correlation optimised correction). The data in this example were aligned based on correlation matching using the base peak chromatogram. Results before and after alignment are shown in Figure 3. The four different alignment algorithms can be combined and repeated. Only the time-axis will be changed, not your data!! |
 |
| |
Figure 3: Alignment Result, Top – original Base Peak Chromatograms, Bottom – BPC chromatograms after alignment (22 seconds). |
| |
Finding Differential Peaks using MS Compare:
MS Compare contains several algorithms to detect differential peaks between two classes or groups of samples. TCM/BCM screening, BioMarker Surface Maps and Direct Peak Picking. |
|
Comparing TCM/ BCM traces
The TCM (Total Chromatogram Mass spectrum) is basically the average mass spectrum of a LC/MS data set (comparable with the Total Ion Current). The BCM (Base Chromatogram Mass spectrum) can be compared to the Base Peak Chromatogram. The advantage of using “mass spectra” is the absence of alignment problems. The user can set the number of TCM units per sample. Typically the chromatogram is divided into 25 sub-sections. The analysis will be performed on each section.
|
|
| |
Figure 4: MS Compare: Differential analysis by comparison of BCM mass spectra. Green lines mark unique m/z values. Bottom - average mass spectra on 14 samples. Top – mass chromatograms for all samples for one of the unique m/z values. |
BioMarker Surface Maps
This method will scan and compare all data sets for unique peaks. MS Compare creates a so-called two-dimensional surface map as shown in figure 5. From the binary map (time vs. m/z) the unique peaks are easily identified. Maps can be created using a number of criteria. In this case full selectivity was used. Alternatives are: absolute or relative differences, t-statistics, Fischer Discriminant value, etc. Plots are interactive; clicking on a unique position will plot the extracted ion currents of all samples in overlay in the top window. |
 |
| |
Figure 5: MS Compare showing unique peaks in a BioMarker Surface Map. |
The map can be converted to a table as shown in Figure 6. Based on this table a number of interactive plots can be created showing all information. The MS Compare module offers different methods to present the results. More advanced (multivariate) techniques like Principal Component Analysis and Clustering are available to get a good overview of the data and results.
|
|
Figure 6: MS Compare – Overview Plots: Top left: plot the Extracted Ion Currents of unique features directly from the result table. Top right: view Extracted Ion Currents or Mass Spectra for any of the selected samples. Bottom left: Profile Plots – intensity versus sample number or intensity versus peak number. Bottom right: Extracted Ion Currents can be directly plotted from the BioMarker Surface Map (overlay or stacked). |
 |
 |
Figure 7: MS Compare – Clustering and PCA |
MS compare is interactive and links directly to the Browser and other modules. At any moment the user will be able to compare a large number of samples based on mass chromatograms or mass spectra, either manually or by using algorithmic techniques. Results can be exported to text files or Excel. |
