COMPUTER-IMPLEMENTED METHOD FOR DETECTING AT LEAST ONE INTERFERENCE AND/OR AT LEAST ONE ARTEFACT IN AT LEAST ONE CHROMATOGRAM

Abstract

A computer-implemented method for detecting at least one interference and/or at least one artefact in at least one chromatogram determined by at least one mass spectrometry device (110) is proposed. The chromatogram comprises a plurality of raw data points. The method comprises the following steps: a) retrieving the at least one chromatogram by at least one processing device (126); b) applying at least one peak fit modelling to the chromatogram by using the processing device (126); c) determining information about residuals of the raw data points by using the processing device (126); d) detecting the at least one interference and/or the at least one artefact by using the processing device (126) by comparing the determined information about the residuals with at least one pre-determined threshold, wherein, if the determined information about the residuals exceed the pre-determined threshold, the at least one interference and/or the at least one artefact is detected.

Claims

1. A computer-implemented method for detecting at least one interference and/or at least one artefact in at least one chromatogram determined by at least one mass spectrometry device, wherein the at least one chromatogram comprises a plurality of raw data points, the method comprising: retrieving the at least one chromatogram by at least one processing device; applying, by the at least one processing device, at least one peak fit modelling to the at least one chromatogram; determining, by the at least one processing device, information about residuals of the raw data points; and detecting, by the at least one processing device, the at least one interference and/or the at least one artefact by comparing the determined information about the residuals with at least one pre-determined threshold, wherein, if the determined information about the residuals exceed the pre-determined threshold, the at least one interference and/or the at least one artefact is detected.

2. The method according to claim 1, wherein retrieving the at least one chromatogram comprises fully automatically retrieving the at least one chromatogram; wherein applying the at least one peak fit modelling to the at least one chromatogram comprises fully automatically applying the at least one peak fit modelling to the at least one chromatogram; wherein determining the information about residuals of the raw data points comprises fully automatically determining the information about residuals of the raw data points; and wherein detecting the at least one interference and/or the at least one artefact comprises fully automatically detecting the at least one interference and/or the at least one artefact.

3. The method according to claim 1, further comprising measuring the at least one chromatogram using the mass spectrometry device.

4. The method according to claim 1, further comprising determining a position of the at least one interference and/or the at least one artefact in the at least one chromatogram.

5. The method according to claim 4, further comprising dividing the at least one chromatogram into at least two sections.

6. The method according to claim 4, wherein the at least one chromatogram is divided into four sections, wherein the at least one chromatogram is divided into a pre-peak section defined between peak start and peak start minus full width at half maximum, an ascending peak section defined between peak start and peak maximum, a descending peak section defined between retention time and peak end and a post-peak section defined between peak end and peak end plus full width at half maximum.

7. The method according to claim 4, wherein determining the position of the at least one interference and/or the at least one artefact in the at least one chromatogram comprises: determining the information about the residuals of the raw data points and comparing the information about the residuals with the at least one pre-determined threshold for each of the sections.

8. The method according to claim 1, wherein the information about the residuals is one or more of the residuals, a mean of the residuals, a median of the residuals, a sum of the residuals, a product of the residuals, and/or an integral of the residuals.

9. The method according to claim 1, wherein applying the at least one peak fit modelling to the at least one chromatogram comprises applying one or more of at least one polynomial interpolation, at least one exponentially modified Gaussian function, at least one Gauss-Newton algorithm, and at least one Fourier-Transformation.

10. The method according to claim 1, further comprising at least one preprocessing step comprising one or more of: selecting at least one region of interest in the at least one chromatogram; selecting at least one predefined retention time interval; performing at least one smoothing by applying one or more of a moving average filter, a Gaussian filter, a discrete wavelet de-noising, a Savitzky-Golay smoothing, a Loess smoothing; and/or performing at least one background subtraction by applying one or more of an asymmetric weighted least squares fit with regularization, a morphological top hat filter, a discrete or continuous wavelet base background determination, and/or a moving average minimum.

11. (canceled)

12. (canceled)

13. A processing system for automatic detection of at least one interference and/or at least one artefact in at least one chromatogram determined by at least one mass spectrometry device, wherein the at least one chromatogram comprises a plurality of raw data points, the processing system comprising: at least one data collector configured to retrieve the at least one chromatogram; at least one fitting unit configured to apply at least one peak fit modelling to the at least one chromatogram; at least one mathematical unit configured to determine information about residuals of the raw data points; and at least one identification unit configured to detect the at least one interference and/or the at least one artefact by comparing the determined information about the residuals with at least one pre-determined threshold, wherein the identification unit is configured to detect the at least one interference and/or the at least one artefact if the determined information about the residuals exceed the pre-determined threshold.

14. (canceled)

15. A mass spectrometry system comprising the processing system of claim 13, and further comprising at least one mass spectrometry device comprising at least one mass filter and at least one detector.

16. One or more non-transitory machine-readable storage media comprising a plurality of instructions stored thereon that, in response to execution by at least processing device, causes a computing system to: retrieve at least one chromatogram, wherein the at least one chromatogram comprises a plurality of raw data points; apply at least one peak fit modelling to the at least one chromatogram; determine information about residuals of the raw data points; and detect at least one interference and/or at least one artefact in the at least one chromatogram by comparing the determined information about the residuals with at least one pre-determined threshold, wherein, if the determined information about the residuals exceed the pre-determined threshold, the at least one interference and/or the at least one artefact is detected.

Description

SHORT DESCRIPTION OF THE FIGURES

[0104] Further optional features and embodiments will be disclosed in more detail in the subsequent description of embodiments, preferably in conjunction with the dependent claims. Therein, the respective optional features may be realized in an isolated fashion as well as in any arbitrary feasible combination, as the skilled person will realize. The scope of the invention is not restricted by the preferred embodiments. The embodiments are schematically depicted in the Figures. Therein, identical reference numbers in these Figures refer to identical or functionally comparable elements.

[0105] In the Figures.

[0106] FIG. 1 shows an embodiment of a mass spectrometry system according to the present invention;

[0107] FIGS. 2A to 2D show representative chromatograms with full separation of an interference (2A), beginning co-elution (2B), strong co-elution (2C), and full co-elution (2D); and

[0108] FIG. 3 shows mean peak fit residual values for section C and D (left y-axis) and relative area ratio (right y-axis) in dependency on mean peak resolution of testosterone and epitestosterone (x-axis).

DETAILED DESCRIPTION OF THE EMBODIMENTS

[0109] FIG. 1 shows, in a highly schematic fashion, an embodiment of a mass spectrometry device 110 according to the present invention The mass spectrometry device 110 comprises at least one mass filter 112 and at least one detector 114. The mass spectrometry device 110 may be part of a mass spectrometry system 111. The mass spectrometry system 111 further comprises a processing system 116. The processing system 116 may implemented as software and/or may be implemented into a processing device 126.

[0110] The mass spectrometry device 110 may be or may comprise at least one liquid chromatography mass spectrometry device. The liquid chromatography mass spectrometry device may be or may comprise at least one high-performance liquid chromatography (HPLC) device or at least one micro liquid chromatography (.Math.LC) device. The liquid chromatography mass spectrometry device may comprise a liquid chromatography (LC) device and a mass spectrometry (MS) device, wherein the LC device and the MS are coupled via at least one interface. The LC device may be configured to separate one or more analytes of interest of the sample from other components of the sample for detection of the one or more analytes with the mass spectrometry device. The LC device may comprise at least one LC column. For example, the LC device may be a single-column LC device or a multi-column LC device having a plurality of LC columns. The LC column may have a stationary phase through which a mobile phase is pumped in order to separate and/or elute and/or transfer the analytes of interest. The mass spectrometry device 110 may be or may comprise a mass analyzer configured for detecting at least one analyte based on mass to charge ratio. The mass filter 112 may be configured for separating components of a sample with respect to their masses. For example, the mass spectrometry device 110 may be or may comprise at least one quadrupole mass spectrometry device. The detector 114 may be configured for detecting incoming particles and for determining the at least one chromatogram. The chromatogram may be a visual result or outcome of a separation process separating components of a sample. The chromatogram may refer to an intensity distribution over time generated during at least one chromatography run. The chromatogram may be or may comprise a diagram with the retention time of the sample components on the x-axis and intensity on the y-axis.

[0111] The chromatogram may comprise at least one peak. The peak may be at least one local maximum of the chromatogram. Specifically, the chromatogram may comprise at least one signal peak. The signal peak may be a peak of an analyte of interest of a sample. The sample may be selected from the group consisting of a physiological fluid, including blood, serum, plasma, saliva, ocular lens fluid, cerebral spinal fluid, sweat, urine, milk, ascites fluid, mucous, synovial fluid, peritoneal fluid, amniotic fluid, tissue, cells or the like. The sample may be used directly as obtained from the respective source or may be subject of a pretreatment and/or sample preparation workflow. The sample may comprise the at least one analyte. For example, analytes of interest may be vitamin D, drugs of abuse, therapeutic drugs, hormones, and metabolites in general.

[0112] The chromatogram comprises a plurality of raw data points. The raw data point may be an entry of the chromatogram and/or a single measurement value of the mass spectrometry device. The raw data point may be preprocessed data such as background subtracted raw data. Specifically, the raw data points may be subjected to a peak fit modelling.

[0113] The processing system 116 may be configured for detection interferences and/or artefacts, wherein the processing system 116 comprises: [0114] at least one data collector 118 configured for retrieving the chromatogram, [0115] at least one fitting unit 120 configured for applying at least one peak fit modelling to the chromatogram, [0116] at least one mathematical unit 122 configured for determining information about residuals of the raw data points; [0117] at least one identification unit 124 configured for detecting the at least one interference and/or the at least one artefact by comparing the determined information about the residuals with at least one pre-determined threshold, wherein the identification unit 124 is configured for detecting the at least one interference and/or the at least one artefact if the determined information about the residuals exceed the pre-determined threshold.

[0118] The retrieving at least one chromatogram may comprise one or more of receiving, downloading, accessing, determining, measuring, detecting, and recording the at least one chromatogram. For example, the chromatogram may be retrieved by downloading and/or accessing the chromatogram from at least one database such as of the detector 114 or of a cloud. For example, the retrieving may comprise measuring the chromatogram using the mass spectrometry device 110. Specifically, the chromatogram may be retrieved by performing at least one chromatography run.

[0119] The peak fit modelling may comprise at least one fit analysis of the chromatogram or at least one region of the chromatogram using at least one fit function. The peak fit modelling may comprise identification and/or detection of a peak of the analyte of interest. The peak fit modelling may comprise one or more of peak detection, peak finding, peak identification, determining peak start and/or peak end, determining of background, determining of basis line and the like.

[0120] The peak fit modelling may comprise applying at least one curve fitting technique to the chromatogram. The raw data points may be used as input values for the peak fit modelling. The peak fit modelling may comprise fitting the raw data points using at least one fit function. The peak fit modelling may comprise applying one or more of at least one polynomial interpolation, at least one exponentially modified Gaussian function, at least one Gauss-Newton algorithm, and at least one Fourier-Transformation For example, the fitting may comprise using at least one fitting function as described in “Mathematical functions for the representation of chromatographic peaks”, Valerio B. Di Marco, G. Giorgio Bombi, Journal of Chromatography A, 931 (2001) 1-30. The method may comprise at least one optimization step comprising determining a best fit function. This so called final peak fit may be used for determining of the information of the residuals.

[0121] The processing system 116 may further comprise at least one preprocessor configured for one or more of selecting at least one region of interest in the chromatogram; selecting at least one predefined retention time interval; at least one smoothing step comprising applying one or more of a moving average filter, a Gaussian filter, a discrete wavelet denoising, a Savitzky-Golay smoothing, a Loess smoothing, at least one background subtraction step comprising one or more of an asymmetric weighted least squares fit with regularization, applying a morphological top hat filter, a discrete or continuous wavelet base background determination, determining a moving average minimum.

[0122] The processing system 116 may allow for advanced detection of artefacts and/or interferences by readouts based on the residuals between the final peak fit and the chromatogram. Based on the peak fit modelling, the residuals may be calculated for each raw data point. The residual may be calculated as a difference between the value of the raw data point at a position of the chromatogram and a value of the final peak fit at said position.

[0123] The information about the residuals may be one or more of the residuals, a mean of the residuals, a median of the residuals, a sum of the residuals, a product of the residuals, an integral of the residuals. The determining of the information about the residuals may comprise determining an absolute value of the residuals before determining mean, median, sum and the like. For example, the mathematical unit 122 may be configured for determining at least one curve of residuals as a function of time and the information about the residuals may be an area under the curve of residuals. For a chromatogram without an interference and/or an artefact the area under the curve would be zero. In case of interferences and/or artefacts the area under the curve would be non-zero. Optionally, the resulting area value may be normalized to the peak area of the fitted analyte.

[0124] The identification unit 124 may be configured for comparing the determined information about the residuals with the at least one pre-determined threshold. The pre-determined threshold may be an arbitrary threshold characterizing a tolerance range for the residuals. For example, the pre-determined threshold may be an allowed maximum for the area under the curve of the curve of residuals. For example, the pre-determined threshold may be 15% of the information about the residuals, preferably 10% of the information about the residuals. For example, the pre-determined threshold may be an allowed maximum for the area under the curve of the curve of residuals normalized to the peak area of the fitted analyte For example, the pre-determined threshold may be < 10 of the information about the residuals, preferably < 5 of the information about the residuals.

[0125] If the determined information about the residuals exceed the pre-determined threshold, the at least one interference and/or the at least one artefact is detected. In case the at least one interference and/or the at least one artefact is detected, the chromatogram and/or the sample may be rejected for further analysis.

[0126] The processing system 116 may be configured for may comprise determining a position of the at least one interference and/or the at least one artefact in the chromatogram. To mimic a manual review, a more detailed declaration of the chromatogram may be provided. The chromatogram may be divided, e.g. by the mathematical unit 122, into more than one section around a detected and fitted peak. For example, the chromatogram may be divided in at least two sections. The information about the residuals, e.g. the area under the curve of the residuals, may be determined individually for each section. Optionally, the resulting area values of the residual sections may be normalized to the peak area of the fitted analyte. The obtained values may represent additional readouts to check/monitor for interferences and/or artefacts.

[0127] The chromatogram may be divided in four sections. Specifically, the chromatogram may be divided into a pre-peak section defined between peak start and peak start minus full width at half maximum (FWHM), an ascending peak section defined between peak start and peak maximum, a descending peak section defined between retention time and peak end and a post-peak section defined between peak end and peak end plus full width at half maximum. For example, outer sections may be defined between peak start and peak start minus FWHM for the pre-peak section and between peak end and peak end plus FWHM for the post-peak section. The pre-peak section and the post-peak section may be of equal range. Additionally or alternatively, the chromatogram may be divided in other four sections. For example, the pre-peak section may be defined between peak start and peak start minus the absolute difference between retention time, i.e. the peak maximum, and peak start. The post-peak section may be defined between peak end and peak end plus the absolute difference between retention time and peak end. The outer sections may be identical for completely symmetric peaks, where peak start and end are positioned with identical distance from the peak maximum. The outer sections may be non-identical for non-symmetric peaks.

[0128] The position of the at least one interference and/or the at least one artefact in the chromatogram may be determined by determining the information about the residuals of the raw data points and comparing the information about the residuals with the at least one pre-determined threshold for each of the sections. Combination of readouts from the different sections may allow a more detailed declaration of the chromatogram what could complement or replace the manual chromatogram review by an expert. Moreover, in contrast to quantifier/qualifier ratios, the readouts may be individual for each mass transition chromatogram and, thus, may not suffer from above-mentioned disadvantages (i), (ii), and (iii) of the known techniques.

[0129] FIGS. 2A to 2D show representative experimental results, in particular chromatograms with full separation of an interference (2A), beginning co-elution (2B), strong co-elution (2C), and full co-elution (2D). Serum samples were spiked with a mixture containing testosterone and epitestosterone as well as testosterone-d3 as internal standard (ISTD). The samples were measured by LC-triple quadrupole(QqQ)-MS in seven different methods and six randomized analytical replicates. Each method consisted of the same MS settings measuring two transitions of testosterone and two of the ISTD testosterone-d3 but variations in the LC gradients. These variations led to different separation powers, i.e peak resolutions, between testosterone and epitestosterone whereby the latter represented the interference as epitestosterone produced signals in both transitions of testosterone in a similar relative extent.

[0130] The raw data points were integrated using an exponentially modified Gaussian fit and residuals between the peak fit and the raw data points calculated for each data point. The chromatogram around the peak were divided into four sections A-D. For example, e.g. as done for the embodiment shown in FIG. 3, section A were a pre-peak section defined between peak start and peak start minus full width at half maximum section B an ascending peak section defined between peak start and peak maximum, i.e. retention time; section C a descending peak section defined between retention time and peak end; and section D a post-peak section defined between peak end and peak end plus FWHM. The pre-peak section and the post-peak section may be of equal range. Additionally or alternatively, as shown in the embodiments of FIGS. 2A to 2D, other four sections may be selected. FIGS. 2A to 2D, the pre-peak section may be defined between peak start and peak start minus the absolute difference between retention time, i.e. the peak maximum, and peak start. The post-peak section may be defined between peak end and peak end plus the absolute difference between retention time and peak end. The outer sections may be identical for completely symmetric peaks, where peak start and end are positioned with identical distance from the peak maximum. The outer sections may be non-identical for non-symmetric peaks.

[0131] Then, the area under the residual curve was calculated individually for each section and normalized to the peak area. For estimating the impact on the result, the area ratio of analyte and ISTD was calculated in addition and set in relation to the area ratio calculated in the method with the highest peak resolution, i.e separation power. For illustration, in FIG. 2 representative chromatograms for certain separation powers are shown with FIG. 2A full separation of an interference, FIG. 2B beginning co-elution, FIG. 2C strong coelution, and FIG. 2D full co-elution. In addition, respective sections A-D as well as a rough visual estimation of their changes are given (“∼” = no change, “↑”increase; “↑↑” =high increase).

[0132] FIG. 3 shows the mean peak fit residual values for section C and D (left y-axis) and relative area ratio (right y-axis) in dependency on mean peak resolution of testosterone and epitestosterone (x-axis). The area ratio (right y-axis), representing the result, was affected at peak resolutions lower than 1.0. This information is usually not known when measuring samples with unknown concentrations. The peak residual section D was affected beginning at peak resolutions lower than 12, showing its maximum at 0.6 and ending at 0.4. Peak residual section C in parallel was affected beginning at peak resolutions lower than 1.0, showing its maximum at 0.4, and ending between 0.4 and 0.0. The peak residual sections A and B remained unaffected for all methods as well as the quantifier/qualifier ratio (data not shown). With certain maximal thresholds for these peak residual section values, such as < 10 for section D and <5 for section C (left y-axis), samples with affected area ratio (right y-axis) could be detected down to peak resolutions below 0.4 between the analyte testosterone and the interference epitestosterone. These interfered samples may have been overlooked by monitoring quantifier/qualifier ratio alone and were usually only detectable by manual chromatogram review by an expert.

[0133] With this described procedure the position of the interference and/or artefact can be estimated, e.g. right-sided or left-sided interference, by combining information of section A and B vs section C and D and/or peak resolution between analyte and interference can be estimated by combining information of section A vs. B or section C vs. D.

TABLE-US-00001 List of reference numbers 110 mass spectrometry device 111 mass spectrometry system 112 mass filter 114 detector 116 processing system 118 data collector 120 fitting unit 122 mathematical unit 124 identification unit 126 processing device