Data-independent acquisition (DIA) promises reproducible and accurate protein quantification across large sample cohorts. The
mass spectrometer typically cycles through many isolation windows covering a broader m/z range of interest. Current methods utilize only about 1-3% of all available ions. In principle, all ions could be utilized by parallel ion storage and sequential release from the TIMS device into a Q-TOF mass analyzer. Here, we asked if the PASEF principle could be transferred to DIA. Methods
Whole-cell proteomes extracted from a human cancer cell line were analyzed via nanoLC coupled to a prototype timsTOF Pro (Bruker). We adapted the instrument firmware to perform data-independent isolation of multiple precursor windows within a single TIMS separation (100ms). We tested multiple schemes for precursor selection window size and placement in the m/z-ion mobility plane. Analysis of the four-dimensional data space has been incorporated into OpenSWATH. For ion mobility-aware targeted data extraction, we used a project-specific library from 48 high-pH reverse-phase peptide fractions acquired with PASEF. Results
As ion mobility and mass are correlated, a large proportion of the peptide ion current can be covered by scanning diagonal lines in the m/z-ion mobility space. We derived multiple diaPASEF acquisition schemes from the density distribution of about 130,000 precursors present in the library. TIMS provides highly precise measurements of collisional cross sections (CCS) with CVs << 1% in technical replicates. After linear alignment, CCS values extracted from the diaPASEF runs deviated < 2% from the library. In triplicate 120min runs of 200ng HeLa digest each, we quantified over 7000 proteins at a 1% FDR. Fragment ion-based quantification was very reproducible with a median CVs of 10% and a pairwise mean Pearson correlation >0.96.
The diaPASEF method captures and utilizes a very large proportion of the available ion current, approaching the ideal mass analyzer.