Data Independent Acquisition (DIA) allows users to gain a comprehensive view of all proteins within their dataset by focusing in on a specific m/z range.
Traditional LC-MS/MS proteomic datasets were collected using Data Dependent Analysis (DDA), also called “top N”. This data collection method would include a survey scan (MS1) which measured the abundance of all the peptides sprayed into the machine. The MS instrument would then attempt to fragment (MS2) the top peaks sequentially, typically this would be in the range of 15-20 peaks. The MS instrument then resumes a new MS1 scan for the next set of ionised peptides and selects the most abundant peaks.
The biggest disadvantage of this approach is that it can lead to a lot of missing values, where peptides are not detected in either the MS1 or MS2 scan. Peptides can be lost either by being too low in abundance to make the top 20 list during the first MS1 scan or are missed entirely as they pass through the machine and are lost while the machine selects for other peptides during the MS2 scan.
DIA allows users to get around these limitations by carrying out MS2 scans directly on all peaks within a predefined m/z range
While DDA selects the “top N” peaks and attempts to fragment them, DIA focuses on a specific m/z range and fragment all precursors (MS2) identified. The scans are conducted in intervals of specific ranges, known as windows. For example, if set up to collect all peptides with a m/z between 500 and 900, the MS scans will be done in 20 windows of 20 m/z segments, starting with 500 – 520, and then 520 – 540, and so on until the entire range from 500 to 900 is covered. After this point the scan will cycle back to the first window, repeating the scan. MS1 scans are conducted periodically in order to help contextualise the data. As a result, the MS2 spectra are very complex and are searched using a spectral library to obtain a very complete understanding of the peptides and proteins that are present in the sample. Since all peptides are fragmented, DIA data can be re-searched with an improved library to obtain improved peptide and protein identification.
The main advantage over DDA is that an MS2 analysis is conducted on every peptide within a given range, and not just the top 20 most abundant, providing a less biased data set with greater sensitivity for low abundance peptides. This technique is especially useful for projects where the user wishes to collect proteomic data on a list of proteins within a known range of m/z ratios. By specifying the range of interest, DIA will focus only on that specific area, filtering out any peptides we aren’t necessarily interested in.
A good analogy to help explain DDA vs DIA, is to consider entering a shop to try and see how many lemons are present. DDA would randomly identify one or two objects from each aisle in the shop, meaning that even if that shop had lemons, it’s possible we won’t get any in our basket just by unfortunate luck. DIA instead would focus on the fruit aisle and take a little bit of everything from each shelf, so if the shop does indeed have lemons, they will definitely be in the basket along with all the other fruit.
• Less sample preparation required
• MS2 scan conducted on all proteins within range
• Not biased towards high abundance proteins
• Fast acquisition compared to DDA
• Spectral library can be researched with an improved library for improved peptide and protein identification
• Requires knowing which range the peptides of interest fall in beforehand
• Any peptides outside of the range will be lost