Intact mass analysis includes two independent deconvolution algorithms for mass spectral data:

  • Xtract, which deconvolves isotopically resolved mass spectra—that is, spectra in which it is possible to distinguish separate peaks for different isotopic compositions of the same component.
  • See Xtract algorithm.
  • ReSpect, which deconvolves isotopically unresolved (or unseparated) mass spectra—that is, spectra in which it is not possible to distinguish the separate peaks for different isotopic compositions of the same component.
  • See ReSpect algorithm.

Whether mass spectra are isotopically resolved or unresolved depends on the resolution of the instrument, the mass of the compounds involved, and the details of the experiment run.

NOTE

Attempting to apply an algorithm to the wrong type of spectral data can lead to unreliable results.

In most cases, the Xtract algorithm fails to identify any components if you apply it to isotopically unresolved spectra as the components do not have isotopic profiles.

If you apply the Respect algorithm to isotopically resolved spectra, it might attempt to identify each isotopic peak as a separate component, rather than an isotopic composition of a single component.

For each of the two deconvolution algorithms, you can specify the deconvolution method:

  • Average over selected retention time: The application deconvolves and averages the user-specified source spectrum over a given retention time (RT) range.
  • Automatic peak detection: The application generates the source spectra using the Parameterless Peak Detection (PPD) algorithm for the automatic peak detection of large molecules.
  • You can use this source spectra method only in Automatic mode.
  • Sliding windows: The application averages spectra over a succession of sliding windows in the retention time range specified by the RT Range parameter. It deconvolves each of these averaged spectra and then merges similar masses to identify components.
  • The sliding windows deconvolution method is a new and powerful approach for identifying components in LC/MS data. For more information, see Sliding windows deconvolution.