One of the most important steps in processing data with LipidView is correctly defining the processing method.
Most processing errors can be traced back to an incorrectly crafted processing method.
It is important to note that due to the extensive isobaric overlap within the lipidome, care needs to be taken to carefully design MS experiments and the methods by which the data is processed in LipidView. LipidView software is not ‘black box’ software and requires the user to take experimental design into account when creating processing methods. When you click on the processing methods icon in the workflow panel, a pull down menu of template methods appears in the top right corner of the main screen area. Find a method that most closely fits the experiment used to acquire the data then click ‘edit methods’. Once you have established a processing method that works for your data, you can save that method using a different file name and can use it to process future data files. The file you created will appear in the template methods pull down menu. Regardless of whether you are creating a new processing method or are using an established one, it is a good idea to always edit your processing method each time you use the software to verify the settings and ensure the processing method is appropriately designed for the data you have acquired.
Processing methods define the parameters and limits by which LipidView software anlayzes your data. The method should reflect the experiment used to acquire the data and settings are instrument and scan-type specific. An important setting in your processing method is the minimum % intensity, which defines the minimum peak intensity required (as a function of the base peak intensity in the spectrum) to be considered for identification. If the background of your data is high, and you do not adjust the minimum % intensity adequately, LipidView will interpret noise as relevant peaks and attempt to assign an identity. This is one of the major causes of false-positive detection. Alternatively, setting the threshold too high will result in missing potentially important, relevant peaks. Under the spectrum/data tab, you need to specify how the sample was introduced to the mass spectrometer, that is by infusion or LC. If you desire, a specific time within the LC run can be targeted. This can be particularly helpful if you know what time a particular class of lipids elutes.
The mass tolerances are key parameters that affect your processed data quality. These values are related to the instrument used and must be adjusted if your instrument is poorly calibrated. A listing of suggested starting points for tolerance settings is given in the attached PDF file. These guidelines are good starting points, however, in most cases, you may need to process your data multiple times to get optimal tolerance settings that minimize false-positive identification. As a suggestion, you may want to start with only one representative data file to minimize the time needed to optimize the tolerance settings.
Under the Processing Methods tab, you must select the method you want the software to use to process your data. By clicking the box for ‘Identify Species’ the software will process your data by accessing the LipidView database and identify masses based on their precursor ion mass. If you have MSMS data, fragments and neutral losses are aligned with precursors to assign peak identities. MSMS data is not required for processing; however, unless a semi-targeted scan mode is used such as precursor ion scans dedicated to a particular lipid class or good chromatographic separation is performed, you may generate high numbers of false positives due to the extensive isobaric overlap of the lipidome. It is recommended you use MSMS data to increase the quality of your data. Other options include using a targeted method. This method can be defined from previously identified results. By clicking the ‘report unidentified peaks box,’ peaks that cannot be identified from the target method or by the identify species method will be available in the results review section. This can be helpful if the peak is a lipid not present in the lipid database.
If you select Identify species as your processing method, you can specify which lipid classes and sub-classes you want to target. As you select lipid classes to include in your search, consider the type of experiment run. For example, if you analyzed a lipid extract looking for molecules that undergo collision-induced dissociation to generate fragment ions with a loss of 141—a loss common to the phospholipid subclass phosphatidylethanolamine—you should only select PE under the glycerophospholipids menu as the species to search for. Due to isobaric overlap, the software will identify peaks from other lipid classes because you are searching by precursor mass alone. By using a semi-targeted scan such as neutral loss of 141, you are using the power of the triple quadrupole instrument to exclude all other lipid classes. The software, however does not know this unless you indicate as such in the processing method. Another consideration is the polarity of the experiment. If you have acquired your data in the positive ion mode, it is very unlikely you will detect anionic phosphatidic acid species. Consequently, this class should be excluded from your processing method. Similarly, searching for neutral lipids such as di- and tri-glycerides is not appropriate in the negative ion mode. For each lipid class or category, there is a details window where you can further define your processing method. For example, for the glycerophospholipids, you can define confirmatory ions (i.e., adducts) as well as define whether a mass that is consistent with an odd fatty acid chain is in fact an odd chain, as is found in bacterial systems, or is an ether-linked lipid, species that are enriched in inflammatory cell types. Once you have defined your processing method, you can save the method for future use.