Optimising Spectronaut search parameters to improve data quality with minimal proteome coverage reductions in DIA analyses of heterogeneous samples

Data independent acquisition has seen breakthroughs that enable comprehensive proteome profiling using short gradients. As the proteome coverage continues to increase, the quality of the data generated becomes much more relevant. Using Spectronaut, we show that the default search parameters can be easily optimised to minimise the occurrence of false positives across different samples. Using an immunological infection model system to demonstrate the impact of adjusting search settings we analysed mouse macrophages and compared their proteome to macrophages spiked with Candida albicans. This experimental system enabled the identification of false positives since Candida albicans peptides and proteins should not be present in the mouse only samples. We show that adjusting the search parameters reduced false positive identifications by 89% at the peptide and protein level, thereby considerably increasing the quality of the data. We also show that these optimised parameters incur a moderate cost, only reducing the overall number of true positive identifications across each biological replicate by <6.7% at both the peptide and protein level. We believe the value of our updated search parameters extends beyond a two-organism analysis and would be of great value to any DIA experiment analysing heterogenous populations of cell types or tissues.