ADAPTIVE INTRINSIC LOCK MASS CORRECTION

Abstract

A method of correcting mass spectral data comprises making calibration measurements of first intrinsic components (A, B, C) at one or more calibration times (t1) using calibrants which have known mass to charge ratio (m/z) values or previously mass measured mass to charge ratio (m/z) values, making a list of second intrinsic components (D, E, F) which are present during more than one acquisition periods, wherein the second intrinsic components have mass to charge ratio (m/z) values that were not present or observed during or close to the one or more calibration times (t1) but which do overlap in time with the first intrinsic components (A, B, C), and utilising the list to calculate a mass or mass to charge ratio (m/z) correction factor for one or more acquisition periods which are not close or adjacent in time to an acquisition period containing a directly calibrated mass to charge ratio (m/z) value.

Claims

1. A method of mass spectrometry comprising: mass analysing a sample; introducing an external or extrinsic lock mass sample or calibrant at a first time t.sub.1; making a first external calibration measurement at or in relation to the first time t.sub.1; recognising one or more intermediate components which are first present or first observed after the first time t.sub.1; and calculating or determining a correction factor based upon at least some of the one or more intermediate components.

2. A method as claimed in claim 1, further comprising: introducing an external or extrinsic lock mass sample or calibrant at a second later time t.sub.2; and making a second external calibration measurement at or in relation to the second time t.sub.2; wherein the step of recognising one or more intermediate components comprises recognising one or more intermediate components which are first present or first observed after the first time t.sub.1 and which are last present or last observed prior to the second later time t.sub.2.

3. A method as claimed in claim 1, wherein the correction factor comprises a mass, mass to charge ratio or time correction factor.

4. A method as claimed in claim 1, further comprising generating a first mass spectral data set and correcting the mass, mass to charge ratio or time of at least a portion of the first the mass spectral data set using the calculated or determined correction factor in order to generate a second mass spectral data set.

5. A method as claimed in claim 1, wherein the one or more intermediate components comprise intrinsic analytes within the sample or intrinsic analyte ions generated from the sample.

6. A method as claimed in claim 1, wherein at least some of the one or more intermediate components which are used to calculate or determine the correction factor comprise first intermediate components which are present or observed during time periods which overlap with the first time t.sub.1 and/or which overlap with the second time t.sub.2.

7. A method as claimed in claim 6, wherein the first intermediate components are directly calibrated.

8. A method as claimed in claim 6, wherein at least some of the one or more intermediate components which are used to calculate or determine the correction factor comprise second intermediate components are not present or which are not observed during a time period which overlaps with the first time t.sub.1 and/or which overlaps with the second time t.sub.2.

9. A method as claimed in claim 8, wherein the second intermediate components are indirectly calibrated.

10. A method as claimed in claim 6, wherein at least some of the first intermediate components and at least some of the second intermediate components are present or observed during time periods which contiguously overlap between the first time period t.sub.1 and the second time period t.sub.2.

11. A method as claimed in claim 1, wherein the one or more intermediate components which are used to calculate or determine the correction factor are present or observed during non-contiguous time periods.

12. A method as claimed in claim 1, further comprising changing one or more voltages applied to one or more ion optical elements in order to compensate for a calculated or determined shift in mass, mass to charge ratio or time.

13. A method as claimed in claim 1, wherein at least some of the one or more intermediate components which are used to calculate or determine the correction factor comprise one or more intermediate components which are not present close or adjacent in time to the first time t.sub.1 and/or the second time t.sub.2.

14. A method of correcting mass spectral data comprising: making calibration measurements at one or more calibration times using calibrants which have known mass to charge ratio (m/z) values or previously mass measured mass to charge ratio (m/z) values; making a list of intrinsic components which are present during more than one acquisition periods, wherein the components have mass to charge ratio (m/z) values that were not present or observed during or close to the one or more calibration times; and utilising the list to calculate a correction factor for one or more acquisition periods which are not close or adjacent in time to an acquisition period containing a directly calibrated mass to charge ratio (m/z) value.

15. (canceled)

16. A mass spectrometer comprising: a device arranged and adapted to introduce an external or extrinsic lock mass sample or calibrant at a first time t.sub.1; and a control system which is arranged and adapted: (i) to mass analyse a sample; (ii) to make a first external calibration measurement at or in relation to the first time t.sub.1; (iii) to recognise one or more intermediate components which are first present or first observed after the first time t.sub.1; and (iv) to calculate or determine a correction factor based upon at least some of the one or more intermediate components.

17. A mass spectrometer as claimed in claim 16, wherein the device is further arranged and adapted to introduce an external or extrinsic lock mass sample or calibrant at a second time t.sub.2 and wherein the control system is further arranged and adapted: to make a second external calibration measurement at or in relation to the second time t.sub.2; and to recognise one or more intermediate components which are first present or first observed after the first time t.sub.1 and which are last present or last observed prior to the second later time t.sub.2.

18. A mass spectrometer as claimed in claim 16, wherein the control system is further arranged and adapted to generate a first mass spectral data set and to correct the mass, mass to charge ratio or time of at least a portion of the first the mass spectral data set using the calculated or determined correction factor in order to generate a second mass spectral data set.

19. (canceled)

20. (canceled)

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0124] Various embodiments will now be described, by way of example only, and with reference to the accompanying drawings in which:

[0125] FIG. 1 shows a representation of a time period during which no conventional lock mass correction is available and wherein horizontal bars represent the elution profiles of analyte ion species with sufficiently good characteristics to be used for calibration purposes;

[0126] FIG. 2 shows the rolling average of ions having a mass to charge ratio of 255 prior to mass correction and illustrates the problem of accurately calibrating ions between two lock mass calibration events;

[0127] FIG. 3 shows a corrected mass to charge ratio of ions having a mass to charge ratio of 255 according to an embodiment and illustrates that a significant improvement in mass calibration may be achieved according to various embodiments; and

[0128] FIG. 4 illustrates a further embodiment and shows a high energy scanning mass filter experiment wherein fragmentation is performed after a first mass filtering step and which illustrates that calibration may be achieved by utilising components which appear in successive experiments or scans such that non-contiguous components may be utilised to provide for mass or mass to charge ratio correction.

DETAILED DESCRIPTION

[0129] Various embodiments will now be described in more detail. According to various embodiments an intrinsic mass to charge ratio calibration method is provided. The method comprises a fully intrinsic method wherein analyte species which are intrinsically or naturally present in an analyte mass spectra are used for lock mass correction.

[0130] The method according to various embodiments differs from other known intrinsic calibration methods in that the method relies upon the detection of analyte ions which may be present for only a portion of an acquisition period rather than background ions which may be present during the whole of an acquisition.

[0131] According to various embodiments the intrinsic species may be present either for a large proportion of the experiment or alternatively the intrinsic species may elute from a separation technique such as a liquid chromatography (“LC”) column over a relatively short period of time.

[0132] According to various embodiments an analytical instrument, mass analyser or mass spectrometer is provided which comprises a chromatography or other separation device arranged upstream of an ion source. The chromatography or other separation device may comprise a liquid chromatography or a gas chromatography device. Alternatively, the separation device may comprise: (i) a Capillary Electrophoresis (“CE”) separation device; (ii) a Capillary Electrochromatography (“CEC”) separation device; (iii) a substantially rigid ceramic-based multilayer microfluidic substrate (“ceramic tile”) separation device; or (iv) a supercritical fluid chromatography separation device.

[0133] The ion source may be selected from the group consisting of: (i) an Electrospray ionisation (“ESI”) ion source; (ii) an Atmospheric Pressure Photo Ionisation (“APPI”) ion source; (iii) an Atmospheric Pressure Chemical Ionisation (“APCI”) ion source; (iv) a Matrix Assisted Laser Desorption Ionisation (“MALDI”) ion source; (v) a Laser Desorption Ionisation (“LDI”) ion source; (vi) an Atmospheric Pressure Ionisation (“API”) ion source; (vii) a Desorption Ionisation on Silicon (“DIOS”) ion source; (viii) an Electron Impact (“EI”) ion source; (ix) a Chemical Ionisation (“CI”) ion source; (x) a Field Ionisation (“FI”) ion source; (xi) a Field Desorption (“FD”) ion source; (xii) an Inductively Coupled Plasma (“ICP”) ion source; (xiii) a Fast Atom Bombardment (“FAB”) ion source; (xiv) a Liquid Secondary Ion Mass Spectrometry (“LSIMS”) ion source; (xv) a Desorption Electrospray Ionisation (“DESI”) ion source; (xvi) a Nickel-63 radioactive ion source; (xvii) an Atmospheric Pressure Matrix Assisted Laser Desorption Ionisation ion source; (xviii) a Thermospray ion source; (xix) an Atmospheric Sampling Glow Discharge Ionisation (“ASGDI”) ion source; (xx) a Glow Discharge (“GD”) ion source; (xxi) an Impactor ion source; (xxii) a Direct Analysis in Real Time (“DART”) ion source; (xxiii) a Laserspray Ionisation (“LSI”) ion source; (xxiv) a Sonicspray Ionisation (“SSI”) ion source; (xxv) a Matrix Assisted Inlet Ionisation (“MAII”) ion source; (xxvi) a Solvent Assisted Inlet Ionisation (“SAII”) ion source; (xxvii) a Desorption Electrospray Ionisation (“DESI”) ion source; (xxviii) a Laser Ablation Electrospray Ionisation (“LAESI”) ion source; (xxix) a Surface Assisted Laser Desorption Ionisation (“SALDI”) ion source; and (xxx) a Low Temperature Plasma (“LTP”) ion source.

[0134] The analytical instrument may further comprise: (i) one or more ion guides; (ii) one or more ion mobility separation devices and/or one or more Field Asymmetric Ion Mobility Spectrometer devices; and/or (iii) one or more ion traps or one or more ion trapping regions.

[0135] According to an approach the accurate masses of the intrinsic species may be stored in a library of known species and/or background ions.

[0136] However, several problems exist with calibration methods using intrinsic species to calibrate mass spectral data. One particular problem is that the ionisation of background ions being used as intrinsic ion species for calibration purposes may be suppressed when analyte species elute. Another problem with the known approach is that the number and type of available background ions may vary during the elution time. This may be due, for example, to applying a liquid chromatography (LC) gradient.

[0137] The problems associated with conventional calibration methods using intrinsic components can become especially acute when the mass scale of the instrument changes over a short timescale. Accordingly, it is important to update the lock mass correction reliably and frequently.

[0138] Various embodiments relate to a fully intrinsic method which may be used instead of or in support of conventional calibration methods as discussed above. The method according to various embodiments may be applied during one or more time periods during an experiment during which no reliable external known lock mass species are available. For example, no reliable known lock mass species may be available either because of the failure (possibly temporary) of another method or because the frequency of introduction of an external lock mass species has been reduced intentionally to avoid interference with the acquisition of analyte data.

[0139] According to various embodiments it is assumed that the mass scale is sufficiently accurate for the application or experiment being performed at the start and/or end of a time period. The accurate mass calibration at the start and/or end of the time period may be due, for example, to a recent instrument calibration or may be due to the use of another lock mass method such as an external lock mass method.

[0140] Accordingly, accurate masses or mass to charge ratios may be assigned to many of the ion peaks which are present in the sample or otherwise observed at one or both endpoints of each time period as shown in FIG. 1.

[0141] With reference to FIG. 1, the two vertical bars are shown which represent or indicate the start t.sub.1 and end t.sub.2 of a time period of interest. The time period may, for example, comprise several or tens of minutes. The calibration of the instrument, mass analyser or mass spectrometer is assumed to be adequate at both the start (t.sub.1) of the time period of interest and also at the end (t.sub.2) of the time period of interest.

[0142] The horizontal bars shown in FIG. 1 correspond with the elution profiles of analyte ion species of interest rather than background ions. The analyte ion species represented by horizontal bars in FIG. 1 may have sufficiently good characteristics such that the ion species may be used to propagate or extend the calibration of the instrument, mass analyser or mass spectrometer to intermediate times between the start (t.sub.1) of the time period of interest and the end (t.sub.2) of the time period of interest.

[0143] It will be apparent that at all times (as exemplified by time points T.sub.1 and T.sub.2 which are shown for illustrative purposes only) at least two or three analyte ion species are eluting and may be used for calibration purposes. In FIG. 1 the ion species are all transient ion species that appear and disappear over the time period of interest. However, it will be appreciated that this is merely for illustrative purposes and there may also be other ion species that may be present across substantially the entire time period of interest which may or may not be utilised.

[0144] Thus, for illustrative purposes only, analyte ion species A, B and C are shown in FIG. 1 as eluting with elution times which overlap with the start t.sub.1 of the time period. Since analyte ion species A, B and C elute with elution times which overlap with the start t.sub.1 of the time period then ion species A, B and C may also be calibrated accurately and hence the time period during which ions may be accurately calibrated may be extended beyond t.sub.1 (towards t.sub.2). This may be performed as a post-processing step.

[0145] In a similar manner, analyte ion species X, Y and Z are shown in FIG. 1 eluting with elution times which overlap with the end t.sub.2 of the time period. Since analyte ion species X, Y and Z elute with elution times which overlap with the end t.sub.2 of the time period then ion species X, Y and Z may also be calibrated accurately and hence the time period during which ions may be accurately calibrated is extended before t.sub.2 (back towards t.sub.1).

[0146] Further analyte ion species such as F, D and E (and U, V and W) which only begin to elute after the first time t.sub.1 and which have finished eluting before the second time t.sub.2 may now be considered.

[0147] It is apparent from FIG. 1 that analyte ion species F elutes with an elution time which overlaps with analyte ion species A and C. Analyte ion species D elutes with an elution time which overlaps with analyte ion species A, B and C. Analyte ion species E elutes with an elution time which overlaps with analyte ion species A and C.

[0148] Accordingly, calibration may be extended to analyte ion species F, D and E based upon already calibrated analyte ion species A, B and C.

[0149] Furthermore, analyte ion species G and H elute with an elution time which overlaps with analyte ion species F, D and E. Accordingly, calibration may be further extended to analyte ion species G and H based upon calibrated analyte ion species F, D and E.

[0150] At the illustrative time T.sub.1 three analyte ion species (G, H and a third unlabeled) are eluting. The elution time of the three analyte ion species overlaps with at least one calibrated analyte ion species F, D and E and demonstrates how the calibration may be further extended.

[0151] Essentially, ions which have been calibrated on the basis of an external lock mass at a first time t.sub.1 may then be used to calibrate a series of later eluting analyte ions such as analyte ion species F, D and E. The later eluting analyte ions F, D and E may in turn be used to calibrate a series of yet later eluting analyte ions such as G and H.

[0152] Accordingly, analyte ions which elute after t.sub.1 may be calibrated based upon a series of previously calibrated analyte ions.

[0153] Similarly, mass spectral data may be post-processed so that ion species which elute before the second later time t.sub.2 may be calibrated and the calibration may be applied working backwards in time towards t.sub.1.

[0154] With reference to FIG. 1, it will be apparent that the corrections to the mass to charge ratios of analyte ion species G, H and T are determined only indirectly by virtue of the overlap of these ions with earlier calibrated analyte ion species F, D and E or later calibrated analyte ion species U.

[0155] Neither analyte ion species F, D, E nor analyte ion species U were present during any calibration period or event t.sub.1, t.sub.2. However, analyte ion species F, D, E and U can still be accurately measured because these analyte ion species are present or are observed at a time which overlaps in time with directly calibrated analyte ion species A, B and C and directly calibrated analyte ion species X, Y and Z which were present during a calibration event t.sub.1, t.sub.2.

[0156] According to various embodiments, the mass correction for a first spectrum M may be determined by comparing it with a previous spectrum M−1.

[0157] When the two scans are compared, N ion peaks may be confidently identified as being common to the two scans.

[0158] According to various embodiments ion peaks which are compromised by interference or saturation may be omitted from this list.

[0159] Accordingly, a list of N mass measurements may be compiled for scan 1:

m.sub.1=(m.sub.11, m.sub.12, m.sub.13 . . . m.sub.1N)   (1)

with associated uncertainties:

σ.sub.1=(σ.sub.11, σ.sub.12, σ.sub.13 . . . σ.sub.1N)   (2)

[0160] Similarly, a list of N mass measurement may be compiled for scan 2:

m.sub.2=(m.sub.21, m.sub.22, m.sub.23 . . . m.sub.2N)   (3)

with associated uncertainties:

σ.sub.2=(σ.sub.21, σ.sub.22, σ.sub.23 . . . σ.sub.2N)   (4)

[0161] Gaussian distributions for the mass measurement errors may be assumed. Furthermore, assuming that the mass shift between the two spectra is a simple constant of proportionality namely m.sub.2≈gm.sub.1 then the maximum likelihood value for g is given by:

[00001]$\begin{array}{cc}g=\frac{{.Math.}_{n=1}^N\frac{m_{1n}{m}_{2n}}{{\sigma }_{1n}^2+{\sigma }_{2n}^2}}{{.Math.}_{n=1}^N\frac{m_{1n}^2}{{\sigma }_{1n}^2+{\sigma }_{2n}^2}}\pm \frac{1}{\sqrt{{.Math.}_{n=1}^N\frac{m_{1n}^2}{{\sigma }_{1n}^2+{\sigma }_{2n}^2}}}& \left(5\right)\end{array}$

[0162] FIG. 2 shows the rolling average of ions having a mass to charge ratio of 255 according to uncorrected mass spectral data. In particular, FIG. 2 shows time-averaged mass measurements from a 50 second infusion experiment of an ion peak having a mass to charge ratio (m/z) of 255 and wherein the mass spectral data is uncorrected for calibration shift.

[0163] The mass to charge ratio measurements were acquired using an orthogonal acceleration Time of Flight instrument or mass analyser.

[0164] To simulate chromatographic replacement of calibration peaks, corrections were calculated using two ion peaks having mass to charge ratios of 785 and 524 in alternating fashion with a period of 5 s. The corrected mass spectral data is shown in FIG. 3 and illustrates both the improvement in mass accuracy which may be obtained according to various embodiments as well as residual statistical behavior.

[0165] Random walk behavior can be reduced by utilizing as many peaks in the data as possible, by linking features together across non-contiguous acquisition time periods where possible and by introduction of additional extrinsic calibrants as often as required.

[0166] According to various embodiments the required frequency of extrinsic calibration events may be determined using predicted ion statistics and predicted random walk behavior.

[0167] Alternatively, the need for an additional calibration event may be determined in real time based on observed ion statistics and predicted random walk behavior.

[0168] An algorithm such as Apex3D™ WATERS CORPORATION® may be used to identify features or regions of interest in the data. A feature may comprise one or more isotopes of any species present in the data for more than one acquisition period. Since the mass to charge ratio of a feature should not vary with retention time, it is possible to relate scans that are not necessarily consecutive. This generalizes the binary approach described above.

[0169] The general problem may be restated as looking for a function g(n) for n1<=n<=n2 where n is acquisition period or scan index and n1 and n2 are the endpoints of the time region in question (so that g(n1) and g(n2) are known).

[0170] With or without the addition of feature detection, imposing smoothness constraints or prior knowledge of the form of g(n) will help to reduce the natural random walk behavior of the approach.

[0171] In particular, restrictions may be imposed on the integrated curvature of g(n) or as an expected change per unit time. This is particularly helpful in post-processing correction, where information from the future (at the end of the time period) can be used to help stabilize lock mass corrections at earlier times. Prior knowledge regarding the expected time dependence of the correction factor could be determined based on a pre-characterisation of the instrument, e.g. using measurements performed using known samples in an environmental chamber or naturally variable environment. Various other approaches would of course be possible.

[0172] The parameters defining g(n) may be determined using a Markov Chain Monte Carlo method or, specifically, nested sampling or one of many other optimization or statistical inference techniques including Bayesian methods. For instance, using a Bayesian approach, a prior probability distribution for the parameters defining g(n) over time may be determined based on knowledge of the correction factor at any direct calibration times, as well as the expected variation of the correction factor over time (e.g. as determined based on a pre-characterisation of the instrument). Any measurements of intrinsic ion species thus represent new data that can be used to update the probability distribution. An updated posterior probability distribution for the parameters defining g(n) can then be calculated in view of the observed intrinsic ion species, and then used to infer representative values for the parameters defining g(n) at any time point(s) of interest, as well as associated error-bars (uncertainties) for the values. As part of this procedure, the unknown mass to charge ratios of the components to be utilized (for example the detected Apex features) may be explored (marginalized).

[0173] The function g expresses the time-dependence of the correction factor (which may be a single parameter or multi-parameter calibration expression such as a polynomial). When the correction factor is a multi-parameter (e.g. linear) correction, prior knowledge relating to the expected time dependence of the correction factor may be given independently for each of the parameters comprising the correction (for example the polynomial coefficients). Alternatively, information may be given in the form of a covariance matrix per unit time with matrix dimensions corresponding to the individual parameters.

[0174] Other non-mass spectral information may be used to help to constrain the behavior of the correction factor g(n), for example temperature measurements that may be correlated in a known way with changes in mass-to-charge measurement.

[0175] In regions where no suitable overlapping species are present, the current lock mass correction may be propagated forwards (or, when post-processing, backwards) unchanged.

[0176] Alternatively, in a post-processing approach, interpolation consistent with the expected time dependence of the correction factor may be used. This interpolation becomes increasingly uncertain at times distant from reliable measurements and this may be reflected in error estimates for the components of the calibration and/or species measured during these periods.

[0177] In a post-processing approach, corrections may be propagated backwards in time in a manner consistent with any prior information regarding the time dependence of the correction factor, and this allows correction of components which first appear prior to the first introduction of an extrinsic lock mass sample or calibrant.

[0178] The technique is not limited to variations on chromatographic timescales. Other embodiments are contemplated wherein there may be measurable changes in mass scale during an ion mobility (or other separator or mass selective ion trap) experiment or scanning or stepping mass filter (or other filter such as a differential mobility spectrometer (“DMS”)) experiment profiled by a sufficiently high-resolution mass spectrometer.

[0179] In these cases, an “acquisition period” may of the order of an individual time of flight experiment or acquisition which may be, for example, of the order of 100 μs. As long as components exist in more than one acquisition period it is possible to use that information to constrain the function g(n) during the separation or scan in question.

[0180] Additionally, the technique is not limited to utilizing features that are present in a contiguous set of acquisition periods. It is possible, for example, that a persistent feature may be unusable for some period of time. This can happen, for example, as a result of a temporary mass interference which could be identified using a change in peak shape or a sudden change in mass inconsistent with the known behavior of the instrument. The feature may also temporarily disappear or may fall below a statistically useful intensity owing to ion suppression or variation in conditions in the ion source.

[0181] Other reasons for repeated appearance and disappearance of a feature include intentionally induced time dependent behavior in the mass spectrometer such as a nested ion mobility separation or scanning mass filter experiment as described above.

[0182] FIG. 4 depicts a high energy scanning mass filter experiment wherein fragmentation is performed after a first mass filtering step and illustrates how the approach according to various embodiments may be extended to non-contiguous mass spectral data.

[0183] Ions may be fragmented using a collision, fragmentation or reaction device. The collision, fragmentation or reaction device may comprise any suitable collision, fragmentation or reaction device. For example, the collision, fragmentation or reaction device may be selected from the group consisting of: (i) a Collisional Induced Dissociation (“CID”) fragmentation device; (ii) a Surface Induced Dissociation (“SID”) fragmentation device; (iii) an Electron Transfer Dissociation (“ETD”) fragmentation device; (iv) an Electron Capture Dissociation (“ECD”) fragmentation device; (v) an Electron Collision or Impact Dissociation fragmentation device; (vi) a Photo Induced Dissociation (“PID”) fragmentation device; (vii) a Laser Induced Dissociation fragmentation device; (viii) an infrared radiation induced dissociation device; (ix) an ultraviolet radiation induced dissociation device; (x) a nozzle-skimmer interface fragmentation device; (xi) an in-source fragmentation device; (xii) an in-source Collision Induced Dissociation fragmentation device; (xiii) a thermal or temperature source fragmentation device; (xiv) an electric field induced fragmentation device; (xv) a magnetic field induced fragmentation device; (xvi) an enzyme digestion or enzyme degradation fragmentation device; (xvii) an ion-ion reaction fragmentation device; (xviii) an ion-molecule reaction fragmentation device; (xix) an ion-atom reaction fragmentation device; (xx) an ion-metastable ion reaction fragmentation device; (xxi) an ion-metastable molecule reaction fragmentation device; (xxii) an ion-metastable atom reaction fragmentation device; (xxiii) an ion-ion reaction device for reacting ions to form adduct or product ions; (xxiv) an ion-molecule reaction device for reacting ions to form adduct or product ions; (xxv) an ion-atom reaction device for reacting ions to form adduct or product ions; (xxvi) an ion-metastable ion reaction device for reacting ions to form adduct or product ions; (xxvii) an ion-metastable molecule reaction device for reacting ions to form adduct or product ions; (xxviii) an ion-metastable atom reaction device for reacting ions to form adduct or product ions; and (xxix) an Electron Ionisation Dissociation (“EID”) fragmentation device.

[0184] According to various embodiments which will be described in more detail with reference to FIG. 4, a mass filter may be provided which may be scanned repeatedly between masses m.sub.1 and m.sub.2 in a manner as shown in FIG. 4. Numerous time of flight mass spectra may be recorded during the scan. The demarcation between spectra is represented by the fine ticks on the x-axis, wherein the x-axis represents time.

[0185] During each scan, parent or precursor ions which are indicated by horizontal lines or elution profiles A, B, C, D, E, F, G are transmitted for a finite period of time which is determined by the resolution of the mass filter as represented by the sloping or slanted lines and the speed of the scan.

[0186] Some parent or precursor ions such as parent or precursor ions C and F are shown as fragmenting to produce a plurality of fragment or daughter ions. The fragment or daughter ions have a range of mass to charge ratios and are present or otherwise observed at the same time as their corresponding parent or precursor ions C and F.

[0187] Some parent or precursor ions may be absent from the high energy data owing to complete fragmentation.

[0188] The extended calibration approach according to various embodiments which has been described above with reference to FIGS. 1-3 may also be applied to mass spectral data obtained in a manner as described with reference to FIG. 4 as long as a sufficient number of components exist in more than one acquisition time period and there is a sufficient proximity of such components to allow mass accuracy to be retained during the scan.

[0189] For example, parent or precursor ions A and B persist for several acquisition time periods and they are both present for several of these time periods.

[0190] Similarly, parent or precursor ions B and C and corresponding fragment ions of parent or precursor ion C persist and overlap and so on.

[0191] The same approach may be applied to low energy filter scanning experiments wherein mainly parent or precursor ions are present, low and high energy ion mobility mass spectrometry experiments and alternating high and low energy MS experiments (MS.sup.E).

[0192] In experiments in which there is significant commonality between successive filter scans or separation cycles (such as the two mass filter scans shown in FIG. 4) the existence of components in successive experiments or scans is a particular example of non-contiguous components as described above.

[0193] Non-contiguous components may be observed in infusion experiments and in experiments in which the scan or separation time is nested inside a slower experiment such as an LC separation.

[0194] The presence of component A at the start of the first and second scans in the embodiment described with reference to FIG. 4 constrains the relationship between g(n) at those times.

[0195] Accordingly, if component A is accurately mass measured during the first high energy scan, then even if it were impossible for some reason to maintain mass accuracy during the first scan it might still be possible to recover mass accuracy at the start of the second high energy scan by effectively using component A as an intrinsic lock mass. A similar approach may be adopted for other common components and fragments.

[0196] The use of Bayesian methods also allows the approach to be extended to any arbitrary, e.g. non-contiguous, spectral data. For example, because the prior probability distribution for the calibration shifts (i.e. the parameters defining g(n)) is based on the expected variation of the correction factor, which is defined continuously over time, inferences at different time points are correlated. Thus, the expected variation of the correction factor can be used to bridge any gaps where there are no or only relatively few intrinsic ion species to cover a larger time period of interest. An updated posterior probability distribution covering all times can therefore be calculated in view of the available knowledge, including any measured intrinsic ion species. The posterior probability distribution can then be used to infer the calibration shift at any time point(s) of interest.

[0197] The applicability of the method according to various embodiments utilises the ability to identify mass spectral features as belonging to a single component in multiple acquisition time periods. This may be simply due to proximity of those acquisition periods in time so that a small mass window can be used to match components with low probability of error. However, there may be other reasons to associate features between acquisition periods. For example, two spectra in an MS imaging experiment may be expected to have significant commonality owing to spatial proximity in the sample being imaged even if a significant time has passed between the acquisition of the spectra. The two spectra may, for example, lie on successive rows of the image.

[0198] Association of species may be probabilistic rather than definitive. This process may be aided by constraints on the type of mass change that may occur between the spectra being associated. Particularly, intrinsic ion species that are present across multiple acquisition periods may be identified by tracking mass spectral peaks across different acquisition periods and associating a number of mass spectral peaks from different acquisition periods with a single ion species when the variation in mass to charge ratio (m/z) for the mass spectral peaks from acquisition period to acquisition period is consistent with an expected mass to charge ratio (m/z) variation over time based on an expected variation in calibration shift. The expected variation in calibration shift could be determined based on a pre-characterisation of the instrument, e.g. using measurements performed using known calibrants in an environmental chamber. Various other approaches would of course be possible.

[0199] Further, if a large number of species are to be used for mass correction, various known statistical methods such as outlier rejection techniques and Markov Chain Monte Carlo sampling methods may be used to reduce or remove the effect of incorrect associations.

[0200] Many different criteria may be used to select appropriate features in the data to use for correction. For example, data that is identified as saturated or interfered with may intentionally be excluded from consideration. As noted above, this selection process may result in components comprising non-contiguous time periods.

[0201] The same approach may also be used to correct measurements other than mass such as drift time in an ion mobility experiment.

[0202] In real time operation, the mass scale may be corrected by changing a voltage or voltages applied to one or more ion optical elements in order to compensate for the calculated shift in mass. The required changes may be determined using a feedback loop.

[0203] Extra terms may be incorporated in the calculation of the correction to prevent any possible instability or wandering in the feedback loop. For example, an optimized Proportional-Integral-Derivative (“PID”) type of control loop may be employed. Such optimised feedback loops may be able to compensate for higher frequency variations than the above “proportional term only” example.

[0204] Additionally, the feedback parameters may be optimised using a machine learning approach.

[0205] Real time calculation and interrogation of the error signal in the loop may be performed on line using Field-Programmable-Gate-Arrays (“FPGAs”).

[0206] Adjusting ion arrival times as opposed to data has the advantage of maintaining mass resolution when multiple spectra are summed, avoiding the need for resampling of the data to maintain the relationship between peak position and time bin locations.

[0207] It will be appreciated that the various embodiments described in detail above are potentially applicable to all Time of Flight instruments or mass analysers. Various embodiments may also be used in conjunction with folded flight path instruments. In particular, the approach according to various embodiments may be implemented on a folded flight path instrument in order to address specific short term mass stability issues with folded flight path instruments.

[0208] Although the present invention has been described with reference to preferred embodiments, it will be understood by those skilled in the art that various changes in form and detail may be made without departing from the scope of the invention as set forth in the accompanying claims.