9
Details
of ZNova Deconvolution Process
1.Perform optional baseline removal (see Figure
10) and data smoothing.
2.Sort remaining m/z, intensity pairs by
intensity (highest – lowest).
3.Estimate the average noise level from the
remaining data points. The user can also apply a relative intensity threshold if
desired.
4.Perform optional data filtering.
5.For each m/z peak, evaluate a range of
candidate charges which is determined
by the input m/z range and the output zero-charge mass range.
6.Compute an intensity-based score which
accounts for all of the peaks that are
part of a contiguous charge state series.
7.Optionally normalize the score based on the
number of charge states observed/predicted.
8.The candidate charge with the maximum score
is assigned the “true” charge
state.
9.If the score of the determined charge state
is above a minimum threshold score,
transform this peak to the zero-charge domain.
10.Repeat steps 5-9 for all other data points in
the spectrum.
For additional information about similar approaches, see references
2-3.