Author(s)

  • Grzegorz Skoraczyński | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Jorge Peinado-Izaguerri (Presenting Author) | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Sebastian Mueller | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • George Rosenberger | Bruker Switzerland AG | Industriestrasse 26, 8117 , Faellanden , Switzerland
  • Oliver M. Bernhardt | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Monika Pepelnjak | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Veronique Laforte | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Roland Bruderer | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Tejas Gandhi | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland
  • Lukas Reiter | Biognosys AG | Wagistrasse 21 , 8952, Schlieren, Switzerland

Abstract

Improving quantification sensitivity is important for maximizing differentially abundant proteins discovery in Data Independent Acquisition (DIA) proteomics experiments. Here, we show how Spectronaut 20 maximizes protein quantification from dia-PASEF data thanks to a refined interference correction algorithm for the ion mobility (IM) dimension as well as an improved IM-dimension peak extraction algorithm. 20 controlled quantification experiments were collected, in which 4 species were combined in various known ratios with one species. Samples were acquired on timsTOF instruments in dia-PASEF mode with different LC gradients. Spectronaut 19 and Spectronaut 20 were used to analyze the datasets in library-free mode. Number of true differentially abundant proteins with 5% error rate (true candidates), number of identifications with CV < 20% (precision) and number of identifications with fold-change error rate < 20% (accuracy) were assessed as indicators of candidate discovery quality. When focusing on E. coli, an organism with the highest concentration ratio and low absolute amount in one condition, Spectronaut 20 improved accuracy by 172%, with precision dropping by less than 6% and number of true candidates increasing by 1%. Summarizing, improvements in the processing of the IM dimension of dia-PASEF data implemented in Spectronaut 20 demonstrated significantly improved quantification accuracy without compromising candidate discovery and with minimal precision drops.