dnoise: Fast Native Data Reduction for Bruker timsTOF
- 1The Scripps Research Institute
Abstract
Bruker timsTOF acquisitions produce dense native .d files whose storage, transfer, and archival become substantial at high throughput. We present dnoise, an open-source Rust tool that removes points directly from timsTOF frames and writes a native-compatible .d directory. dnoise retains ions that form coherent streaks across the ion-mobility dimension and applies acquisition-aware gates to signal that cannot be selected for fragmentation. On a three-species benchmark spanning ddaPASEF and diaPASEF at 5- and 15-minute gradients, default MS1-only denoising reduced the frame binary by 35 to 53%. Label-free quantification accuracy was preserved in both modes. ddaPASEF peptide-spectrum-match, peptide, and protein-group counts were unchanged, as expected with the searched MS/MS spectra untouched, and diaPASEF precursor and protein-group counts changed only slightly. Every tested processing run completed in 69 seconds or less on the benchmark workstation. Optional MS/MS denoising produced greater reduction but sacrificed several percent of identifications. Thus, a substantial fraction of native timsTOF frame data can be removed with little analytical change.
Related
- When Roads Remove Trees, Conservation Must Move Fast: A Rescue Translocation Model for Keralam’s Native Epiphytic Orchids
- Spontaneous native–non-native tree mixtures worldwide: a systematic map of research patterns, with knowledge gaps and management implications
- Flavonoid-mediated plant-microbiome interactions shape competitive outcomes between native and non-native mangroves
- Broadband Rheometry of Native Soft Tissues via i-Rheo-Indent
- Physicochemical properties of native and extruded maize flours in the presence of animal proteins
BibTeX
@misc{garrett2026dnoise,
title = {{dnoise: Fast Native Data Reduction for Bruker timsTOF}},
author = {P. T. Garrett and J. K. Diedrich and J. R. Yates},
howpublished = {bioRxiv},
year = {2026},
month = aug,
doi = {10.64898/2026.08.27.747603},
url = {https://www.biorxiv.org/content/10.64898/2026.08.27.747603v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - dnoise: Fast Native Data Reduction for Bruker timsTOF
AU - P. T. Garrett
AU - J. K. Diedrich
AU - J. R. Yates
PY - 2026
DA - 2026/08/27
PB - bioRxiv
DO - 10.64898/2026.08.27.747603
UR - https://www.biorxiv.org/content/10.64898/2026.08.27.747603v1
AB - Bruker timsTOF acquisitions produce dense native .d files whose storage, transfer, and archival become substantial at high throughput. We present dnoise, an open-source Rust tool that removes points directly from timsTOF frames and writes a native-compatible .d directory. dnoise retains ions that form coherent streaks across the ion-mobility dimension and applies acquisition-aware gates to signal that cannot be selected for fragmentation. On a three-species benchmark spanning ddaPASEF and diaPASEF at 5- and 15-minute gradients, default MS1-only denoising reduced the frame binary by 35 to 53%. Label-free quantification accuracy was preserved in both modes. ddaPASEF peptide-spectrum-match, peptide, and protein-group counts were unchanged, as expected with the searched MS/MS spectra untouched, and diaPASEF precursor and protein-group counts changed only slightly. Every tested processing run completed in 69 seconds or less on the benchmark workstation. Optional MS/MS denoising produced greater reduction but sacrificed several percent of identifications. Thus, a substantial fraction of native timsTOF frame data can be removed with little analytical change.
ER -