Median filter for liquid chromatography-mass spectrometry data
View Patent ↗High-intensity, spiked noise is reduced in chromatography-mass spectrometry data by applying a nonlinear filter such as a moving median filter to the data. The filter is applied to individual mass chromatograms, plots of ion abundance versus retention time for each detected mass-to-charge ratio, and the filtered chromatograms are combined to form a filtered total ion current chromatogram. Standard linear filters are not effective for reducing noise in liquid chromatography-mass spectrometry (LC-MS) data because they assume a normal distribution of noise. LC-MS noise, however, is not normally distributed.
1. A method for characterizing a chemical sample, comprising:
acquiring a series of mass spectra of said sample;
generating a set of discrete data points based on the mass spectra, said data points comprising a plurality of chromatograms;
selecting a data filter window size;
centering said data filter window on one or more particular data points of said set of data points; and
replacing said particular data point with the median of all points within said window to produce a filtered mass chromatogram.
2. The method of claim 1 , further comprising:
applying a component selection algorithm to said filtered mass chromatogram.
3. The method of claim 1 , wherein said discrete data points correspond to integer mass values.
4. The method of claim 1 , wherein said discrete data points correspond to non-integer mass values.
5. The method of claim 1 , further comprising:
combining a plurality of filtered mass chromatograms into a total ion current chromatogram.
6. The method of claim 5 , wherein said combining comprises summing a plurality of intensities associated with said filtered mass chromatograms.
7. The method of claim 5 , wherein said combining comprises thresholding said mass spectra using said plurality of filtered mass chromatograms.
8. The method of claim 1 , wherein said data window filter size is selected based on at least one of:
(a) being not wider than a typical data peak width;
(b) a scan rate at which said generating is performed;
(c) data characteristics obtained by analyzing subsets of points in said plurality of mass chromatograms; and
(d) a predetermined knowledge of instrument conditions.
9. A program storage device accessible by a processor, tangibly embodying a program of instructions executable by said processor to perform method steps for a chemical sample analysis method, said method steps comprising:
acquiring a series of mass spectra of said sample;
generating a set of discrete data points based on the mass spectra, said data points comprising a plurality of chromatograms;
selecting a data filter window size;
centering said data filter window on one or more particular data points of said set of data points; and
replacing said particular data point with the median of all points within said window to produce a filtered mass chromatogram.
10. The program storage device of claim 9 , said method steps further comprising:
applying a component selection algorithm to said filtered mass chromatogram.
11. The program storage device of claim 9 , wherein said discrete data points correspond to integer mass values.
12. The program storage device of claim 9 , wherein said discrete data points correspond to non-integer mass values.
13. The program storage device of claim 9 , said method steps further comprising:
combining a plurality of filtered mass chromatograms into a total ion current chromatogram.
14. The program storage device of claim 13 , wherein said combining comprises summing a plurality of intensities associated with said filtered mass chromatograms.
15. The program storage device of claim 13 , wherein said combining comprises thresholding said mass spectra using said plurality of filtered mass chromatograms.
16. The program storage device of claim 9 , wherein said data window filter size is selected based on at least one of:
(a) being not wider than a typical data peak width;
(b) a scan rate at which said generating is performed;
(c) data characteristics obtained by analyzing subsets of points in said plurality of mass chromatograms; and
(d) a predetermined knowledge of instrument conditions.
17. A method of characterizing a chemical sample, comprising:
obtaining a chromatogram from a series of mass spectra of said sample, wherein said mass spectra are generated by chromatography and mass spectrometry, and wherein said chromatogram comprises signal data and noise data; and
applying a non-linear filter to said chromatogram to remove at least some of said noise data, thereby producing a filtered chromatogram.
18. The method of claim 17 , wherein said chromatogram is a total ion chromatogram and said filtered chromatogram is a filtered total ion chromatogram.
19. The method of claim 18 , further comprising generating individual chromatograms from said total ion chromatogram, wherein said non-linear is applied to said individual chromatograms.
20. The method of claim 17 , wherein said chromatography is liquid chromatography.