Skip to content

Shift, trim, and bin traces (Other detectors)

Info

To process data from other detectors, the Other detector service of mzmine is required. This feature is included in the mzmine PRO subscription. Academic program and other users can reach out to inquire access options.

Commonly, a shift in the observed retention time is observed between the MS detector and other detectors due to different lengths in flow path.

Sudden changes in background intensity, e.g., by changes in solvent composition, may lead to subpar performance of the baseline correction. In such cases, it may be useful to trim the traces of other detectors.

As other detectors typically have a higher acquisition rate than mass spectrometers, it may be useful to bin the detected intensities/absorptions together to achieve a similar acquisition rate for better cross-correlation and processing performance.

Parameters

shift_traces.png

Raw data files

Select the MS data files to shift the selected traces of.

Trace selection

Select the specific traces you want to shift. Typically, it is not necessary to shift all existing traces, but only the ones you want to correlate with the MS, e.g., the UV (=Absorption) traces.

For a detailed description of all sub-parameters see trace selection parameter.

RT shift (manual/auto)

Sets the retention time offset that is added to each selected other detector trace. Positive values shift the trace to later retention times, negative values to earlier retention times.

  • Manual applies the entered offset (in minutes) directly. Default: 0.
  • Auto determines one global offset automatically from the selected raw data files by matching peaks of the binned other detector trace against peaks of the MS base peak chromatogram. See Algorithm for details.

Tip

Use Auto when the same instrument/method is reused across files, so a single shared offset is appropriate. For files with diverging chromatography, run the module per file or fall back to Manual.

Trim RT range (Optional)

Retention time range to keep. Trimming is applied after shifting.

Bin width (manual/auto) (Optional)

Number of consecutive data points to bin together.

  • Manual uses the entered bin width.
  • Auto sets the bin width so that the resulting other detector acquisition rate is roughly four times the MS acquisition rate (computed per file as floor(other_rate / ms_rate) / 4, with a minimum of 1).

Binning is applied after shifting and trimming.

Algorithm

Automatic RT shift

When Auto is selected for the RT shift, one global offset is calculated across all selected files using the already binned other detector traces:

  1. Per-file peak detection. The binned other detector trace and the MS base peak chromatogram are independently resolved into peaks using a wavelet-based peak picker.
  2. Unambiguous peaks only. Peaks that have a direct neighbouring peak are discarded, since their apex matching would be ambiguous.
  3. Apex-based candidate shifts. Each remaining other detector peak is paired with MS peaks within an apex tolerance window to generate candidate shifts.
  4. Shape correlation validation. Each candidate shift is validated by Pearson correlation of the chromatographic shapes of the matched peaks. Matches require at least 5 correlated data points and a Pearson r ≥ 0.8.
  5. Per-file shift. Validated matches are clustered and the dominant shift for the file is selected. Files with fewer than 3 validated matches are skipped.
  6. Global shift. Per-file shifts from all valid files are combined into a single global offset that is applied to every selected file.

If no file yields a valid shift, the module fails with an error. Otherwise, the preview reports the determined shift together with the number of contributing files and highlights the validated peak matches as filled regions on the shifted trace:

auto_shift_preview.png

Warning

The automatic shift requires real LC-style peaks in both the MS BPC and the other detector trace. Traces dominated by baseline drift, single broad humps, or detector saturation will typically be skipped because they yield too few unambiguous, shape-correlated matches.