The combination of short liquid chromatography (LC) gradients and data-independent acquisition (DIA) by mass spectrometry (MS) has proven its huge potential for high-throughput proteomics. However, the optimization of isolation window schemes resulting in a certain number of data points per peak (DPPP) is understudied, although it is one of the most important parameters for the outcome of this methodology. In this study, we show that substantially reducing the number of DPPP for short-gradient DIA massively increases protein identifications while maintaining quantitative precision. This is due to a large increase in the number of precursors identified, which keeps the number of data points per protein almost constant even at long cycle times. When proteins are inferred from its precursors, quantitative precision is maintained at low DPPP while greatly increasing proteomic depth. This strategy enabled us to quantify 6018 HeLa proteins (>80 000 precursor identifications) with coefficients of variation below 20% in 30 min using a Q Exactive HF, which corresponds to a throughput of 29 samples per day. This indicates that the potential of high-throughput DIA-MS has not been fully exploited yet. Data are available via ProteomeXchange with identifier PXD036451.
Keywords: SPEED; data points per peak; data-independent acquisition; isolation window; predicted spectral library; quantitative proteomics.