Processing of spatially resolved ion spectroscopy measurement signal data for determining molecular content fractions in two-dimensional samples

Abstract

The invention relates to a method for processing ion spectroscopy measurement signal data recorded spatially resolved on a two-dimensional sample, the method comprising: - providing measurement signal data with a plurality of measurement signal histograms, wherein the histograms contain measurement signal tuples with an intensity dimension (J), a mass dimension (m) and a collision cross section dimension (σ) or quantities derived therefrom; - specifying a first and a second selection of ion species of the sample, whose presence in the histogram can be detected and distinguished using the collision cross section dimension or an alternative measurement indicator; - determining a limited area (A fin，x，y ) and calculate the different contents to form a spatially resolved content score (G x，y );‑content fraction (G x，y ) labeling samples. The present invention also relates to a method for acquiring and processing ion spectrum measurement signal data, and an ion mobility-mass spectrometer.

Description

Technical Field

[0001] The invention relates to a method for processing spatially resolved ion spectroscopic measurement signal data recorded on a two-dimensional or flat sample, in particular those data obtained and separated into analytically independent collision cross section dimensions (σ) and mass dimensions (m). Background Art

[0002] The prior art is explained below with reference to specific aspects. However, this should not be construed as a limitation of the scope of the present disclosure. Useful further developments and modifications known from the prior art may also be applied above and beyond the relatively narrow scope of this introduction, and will be apparent to those skilled in the art after reading the following disclosure.

[0003] MALDI mass spectrometry

[0004] Spatially resolved or imaging MALDI mass spectrometry (MALDI MSI, matrix-assisted laser desorption and ionization mass spectrometry imaging) is an analytical method particularly useful for analyzing biological tissue samples. Using this method, a continuous tissue sample is usually scanned point by point using a laser, and a data set is created containing the complete mass spectrum of each measurement point; the mass spectrum can also be calculated as a total spectrum of many individual spectra from the same point, because the signal-to-noise ratio is better. It is therefore possible to generate spatially resolved molecular features of biomolecules, such as lipids, peptides, proteins and polysaccharides, directly from tissue sections, see, for example, US 5,808,300, which is a prior disclosure of this imaging method.

[0005] In preparation for a MALDI MSI measurement, the tissue sample is coated with a special matrix solution. During the measurement, the crystalline matrix clusters absorb the energy of the laser beam and are thus extracted from the tissue. In the process, they also entrain biomolecules from the tissue that had been embedded in the matrix clusters during the crystallization process. The biomolecules acquire an electrical charge in the process, i.e. are ionized. The charged molecules thus extracted are subsequently separated according to their mass-to-charge ratio m / z and detected in the mass spectrometer. Therefore, in addition to the analyte molecules from the tissue, matrix molecules and matrix clusters are always detected. However, it is desirable that the proportion of matrix is ​​as small as possible, since it is an artifact of the sample preparation and ionization method and contains hardly any meaningful analytical information about the sample itself.

[0006] The magnitude of the matrix fraction depends on several factors, such as the ambient temperature, pressure and humidity at the time of matrix application and the nature of the sample being analyzed. In addition, many tissue sections have pores of varying sizes, where it is of course impossible to measure any analyte molecules, so that the matrix fraction is correspondingly high.

[0007] Ion Mobility Spectrometry-Mass Spectrometry

[0008] In a mass spectrometer, molecules are separated simply according to their mass-to-charge ratio, m / z. Many ionization methods (e.g., MALDI) produce singly charged ions (z=1), so the concept of mass (m) can be used simply. Therefore, it is usually not possible to distinguish different molecules of the same or nearly the same mass in a mass spectrometer.

[0009] In ion mobility spectrometry-mass spectrometry (IMS-MS), charged molecules go through another stage before arriving at the mass spectrometer. In this stage, gas and electric field are used to separate charged molecules according to their collision cross section (σ). This means that charged molecules are actually introduced into the mass spectrometer according to their collision cross section classification. The more common parameter mobility (K) is usually used, which is inversely proportional to the collision cross section. Instead of a one-dimensional mass spectrum, a two-dimensional collision cross section-mass signal histogram can be obtained for each measurement point, which usually contains three dimensions: collision cross section σ (or mobility K), mass-to-charge ratio m / z (or simplified to mass m) and intensity (J). See Figure 1 of the prior art, in which the vertical drift time axis (mobility drift time) is proportional to the collision cross section, and the intensity is drawn in color scale in the original image and in gray scale in the image reproduced and presented here.

[0010] All detected charged molecules generate signal peaks at specific positions in the histogram according to their mass and collision cross section. Mass and collision cross section are roughly related to each other (heavier molecules are generally larger), but the precise relationship depends on the class of molecule in question. Therefore, when looking at many different biomolecules, the collision cross section of a light molecule may be larger than that of a heavy molecule, and vice versa. For lipid class molecules, which are important in many applications, but also for other molecules, it can also be assumed that they follow a specific curve in the collision cross section / mass plot, such as shown by the trend line in Figure 1.

[0011] A paper by Kamila Chughtai et al. (Journal of Lipid Research Vol. 54, 2013 333-344) describes the discovery, identification and spatially resolved visualization of several lipid species in a human breast tumor xenograft model using MALDI-IMS-MSI.

[0012] A method for visualizing and quantifying different relative signal contributions was presented at the Imaging Mass Spectrometry Conference (OurConVII), held at the Grand Palais in Saint-Malo, France, from October 28 to 31, 2019. This method is based on mass defect filtering, i.e., statistical models based on the expected masses of peptides, glycans and matrix clusters (excluding collision cross sections, which are not included in the data). By comparing the molecular background signals of pure mass spectra with these models, local mass shifts can be estimated with high accuracy and each spectrum is decomposed into signal components, which can be assigned to different classes of molecules. For each pure mass spectrum, a signal quality score (SQS) from 0% to 100% is calculated, which represents the relative contribution of peptide / glycan molecules above the matrix background. Visualization of these scores makes it possible to identify areas with low signal quality and to assess the overall data quality and homogeneity.

[0013] It is necessary to be able to quantitatively evaluate the spatial distribution of selected molecules of interest in the analyzed two-dimensional or flat sample, which spatial distribution is different from other mass signals, especially background measurement signals, and the evaluation is based on the ion mobility spectrometry-mass spectrometry data obtained thereby. By reading the following disclosure, other purposes that can be achieved by the present invention will become apparent to those skilled in the art. Summary of the invention

[0014] The present invention relates to a method for processing ion spectrum measurement signal data recorded spatially resolved on a two-dimensional sample. Preferably, a wide two-dimensional sample is used, such as a flat tissue slice, or an array of separately prepared samples is used, such as a tissue homogenate preparation for local application. This two-dimensional sample can be treated with a matrix material to perform matrix-assisted laser desorption (especially the entire surface). It is particularly worth mentioning that the MALDI process, in which the same laser emission can achieve desorption and ionization. However, it is also possible to trigger desorption with a first laser emission, and then to subsequently ionize the desorbed molecules by means of a second laser emission that is coordinated in time with the first laser emission. It is also considered that there are no matrix treatments for generating measurement signal data, which are evaluated by the proposed method, such as measurement signal data generated by bombarding the sample with primary ions using SIMS (secondary ion mass spectrometry), or by exposing the sample to a solvent sprayed while applying a voltage gradient using DESI (desorption electrospray ionization).

[0015] The measured signal data having a plurality of measured signal histograms are provided. The measured signal histograms are assigned to a limited area (A) of the two-dimensional sample by means of two position coordinates (x, y). fin，x，y ), which is smaller than the total area of ​​the two-dimensional sample (A total), the histogram further comprising a measurement signal tuple having an intensity dimension (J) or a quantity derived therefrom, a mass dimension (m) or a quantity derived therefrom and a collision cross section dimension (σ) or a quantity derived therefrom. The two position coordinates (x, y) in the measurement signal histogram can be supplemented with a depth dimension (t) to form a triple coordinate (x, y, t), for example, when the desorption beam successively probes different depths of a two-dimensional sample in several ablation layers in order to measure the signal content. Furthermore, the measurement signal tuple can (i) be supplemented with a charge dimension (z) or a quantity derived therefrom (e.g. m / z), and / or include a mobility dimension (K), the latter being derived from the collision cross section (σ). For the sake of clarity, it should be emphasized here that, in the context of the present disclosure, the raw data on which the above-mentioned physical parameters are based may also be considered as derived quantities, such as the flight time derived from a time-of-flight mass spectrometer and the voltage amplitude derived from an ion detector, since these are representative quantities of physical parameters that cannot be directly observed (e.g., flight time => mass m or m / z; voltage => intensity J) and can usually be converted into this form by precisely specified calculation rules.

[0016] In addition, a first and second selection of ion species are specified for the two-dimensional sample, the presence of these selected ion species in the measurement signal histogram is detectable and can be distinguished using the collision cross section dimension (σ) or a quantity derived therefrom (e.g., mobility K). If necessary, the area to be selected can be manually defined in the graphical representation of the measurement signal histogram generated by the computer. The first selection of ion species most preferably includes species with high analytical significance, and the second selection of ion species most preferably includes species with low analytical significance. The first selection of ion species can include, for example, proteins, peptides, glycans and / or lipids in the two-dimensional sample. The second selection of ion species can in particular include charged atoms or molecules and / or clusters thereof generated by the sample preparation method and / or the ionization method. In principle, different biomolecules in the two-dimensional sample can also be specified as the first and second selections relative to each other. Thus, for example, the measurement signals of lipids, glycans and / or peptides / proteins can be compared with each other, or they can be regarded as ratios relative to each other. It is also conceivable to use different charge states (z) of molecules of different classes and / or in molecules of a single class for the first and second selections of ion species.

[0017] The first selection of ion species can be specified by adding several measurement signal histograms into a superimposed measurement signal histogram and identifying a meaningful (possibly dominant) part of the measurement signal tuple entries in the superimposed measurement signal histogram. This can be done by regression analysis, for example by logarithmic regression or logarithmic Radon transformation. In at least one dimension of the measurement signal histogram, the meaningful part of the measurement signal tuple entries can be used to distinguish between the first selection and the second selection of ion species. For example, separate regions in the collision cross section-mass plane (or a plane with corresponding derived quantities) of the superimposed histogram can be identified, which regions mainly have measurement signal contents of different origins, for example lipid and / or peptide regions on the one hand and matrix background regions on the other hand.

[0018] A regression analysis can be designed so that according to the equation σ(m)≈Cm k (power law) is used to explore the correlation between the collision cross section (σ) and the mass (m) of a molecule, where C is a molecule-dependent scaling factor and k is a molecule-dependent exponent. This correlation between the collision cross section (σ) and the mass (m) can be easily converted into quantities derived therefrom. The constants C (scaling factor) and k (exponent) depend on the class of the molecule as well as the instrument and acquisition parameters. Independent power laws can be applied to many different classes of molecules (e.g., lipids, peptides, glycans). Zhiwei Zhou et al. illustrate this correlation for lipids (“LipidCCS: Prediction of Collision Cross-Section Values ​​for Lipids with High Precision To Support Ion Mobility-Mass Spectrometry-Based Lipidomics”. Analytical Chemistry, 2017, 89(17): 9559-9566).

[0019] The superimposed measurement signal histogram can be calculated by the position-independent sum of multiple measurement signal histograms. In addition, the position-independent sum can only take into account such measurement signal histograms, in which the measurement signal tuple entries of at least one dimension (i) exceed a predetermined threshold, (ii) are below a predetermined threshold, or (iii) are within a predetermined value range. For example, a minimum intensity, or a maximum intensity or an intensity range can be specified. Generally, intensity, collision cross section and / or mass range (or the range of corresponding derived quantities) can also be selected as preferred considerations. In this way, mass signal parts that provide little information from the outset or are considered to be uninformative can be masked.

[0020] The method may also be designed to transform at least one measurement signal tuple entry of each measurement signal histogram, for example in the intensity dimension (J), before summing, so that the measurement signal tuple entries of a first predetermined range are weighted disproportionately with respect to a second predetermined range.

[0021] In a limited area (A fin，x，y ) in the measured signal histogram, and the spatially resolved content from the first selected ion species and the spatially resolved content from the second selected ion species are determined, and the various contents are calculated to give a spatially resolved content fraction (G x，y In one example, the spatially resolved content fraction (G x，y ) can be calculated as follows

[0022]

[0023] where S represents the quantity of label i for which the corresponding measurement signal tuple entry of a single measurement signal histogram (e.g., m i , σ i The method of calculating the measured signal from the first selected ion species is described in detail herein. The method of calculating the measured signal from the first selected ion species is described in detail herein. The method of calculating the measured signal from the first selected ion species is described in detail herein. The method of calculating the measured signal from the first selected ion species is described in detail herein. The method of calculating the measured signal from the first selected ion species is described in detail herein. The method of calculating the measured signal from the first selected ion species is described in detail herein.

[0024] However, in further embodiments, a calculation method is also possible which directly compares the first and second selections of ion species, for example according to the following equation:

[0025]

[0026] This calculation method provides content scores that represent the actual content ratio of each individually selected ion species and are therefore easier to interpret.

[0027] Furthermore, it is conceivable and intended to define a second selection of ion species such that it does not include all ion species that are not assigned to the first selection of ion species. For example, a second signal channel or range can be determined for a specific second signal portion in the collision cross section-mass plane (or in the plane of the corresponding derived quantity). This approach is useful in such a measurement signal histogram, that is, it contains signal portions of several different substance classes, such as lipids and / or peptides and / or polysaccharides, which are candidates for meaningful signal portions in principle as biomolecules, as well as matrix clusters or other background ion species, which are essentially meaningless. The content fraction can be calculated as before by:

[0028] or

[0029] Where S a and S b respectively refer to the quantities of the labels i for which the corresponding measurement signal tuple entries of the individual measurement signal histograms (e.g., m i , σ i The first or second selection of ion species (or the amount derived therefrom) is assigned to each. Furthermore, the advantage of the second calculation method described above is that, for example, when the first and second selection of ion species contain different biomolecules in a flat sample and their ratio is calculated, the content fractions can simply be reversed, thus producing an informative result.

[0030] Finally, spatially resolved content fractions (G x，y ). After labeling, the user can be presented with an image of a two-dimensional or flat sample in which a single limited area (A fin，x，y ) with a specified content fraction (G x，y ) can be visually encoded, for example by a color scale or a gray scale. The predetermined value range of the content fraction can be classified as ion spectroscopic measurement signals from outside the two-dimensional sample, which can reasonably be assumed to contain no information of any analytical use, such as where a tissue section is cut or has holes, or in the area between multiple individual samples applied to a sample holder. Additionally or alternatively, the subsequent evaluation (i) can only take into account the content fraction (G x，y ) is within a predetermined range of values, for example, above a minimum value, below a maximum value, or within a limited region within a range of values ​​(A fin，x，y ) and / or (ii) using the content fraction (G x，y ) as a weighting factor. For example, the weighting factor can be used for adaptive noise suppression in subsequent data analysis.

[0031] Furthermore, the present invention relates to a method for acquiring and processing ion spectrum measurement signal data, wherein (i) ion spectrum measurement signal data has been acquired by means of an ion mobility spectrometer-mass spectrometer, and (ii) the method as described above is performed to process the acquired data.

[0032] The invention also comprises an ion mobility spectrometer-mass spectrometer having a computing and / or control unit, which is designed and configured to carry out one of the above-described methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The present invention may be better understood with reference to the following drawings, wherein the elements are not necessarily drawn to scale, but are primarily intended to illustrate the principles of the invention (mostly schematically).

[0034] Figure 1 shows a collision cross section versus mass signal histogram with highlighted trend lines for molecular classes from a prior art paper by Shelley N. Jackson et al. (J. Mass Spectrometry, 2007 Aug; 42(8); 1093-1098; Figure 1 and text caption).

[0035] FIG2 depicts a schematic diagram of the adaptability of an ion mobility spectrometer-mass spectrometer from a prior art paper by Jeffrey Spraggins et al. (Analytical Chemistry 2019, 91, 22, 14552-14560; FIG1 thereof), with which a collision cross section-mass signal histogram (e.g., position coordinates x, y; signal tuples m, σ, J) can be obtained. For example, a special method for mobility separation (parallel cumulative trapped ion mobility spectrometry-PATIMS) is described in more detail in the applicant's published patent application EP 3 054 473 A1 or US2016 / 0231275A1.

[0036] Figure 3 Schematic representation of a flat sample in the form of a tissue slice, here mimicking a thin slice of a mouse brain, is shown for illustrative purposes, once with (lower left) and once without (upper right) a square grid of points or pixels for ion spectroscopy images.

[0037] Figure 4 An example of a collision cross section versus mass signal histogram for a MALDI IMS-MSI measurement is shown. The mass (or more precisely, the mass-to-charge ratio, expressed as m / z) is plotted on the horizontal axis; the collision cross section (expressed as the derived quantity 1 / K0) is plotted on the vertical axis.

[0038] Figure 5 is a logarithmic representation of the superimposed collision cross section-mass signal histogram showing the regression line (solid white line) and signal channels (dashed lines) of the main signal components of interest.

[0039] Figure 6 Shows Figure 5 Logarithmic collision cross section in - Radon transform of the mass signal histogram. The projection angle is plotted on the horizontal axis and the distance from the center of the logarithmic histogram is plotted on the vertical axis. The white cross identifies the location of the maximum intensity.

[0040] Figure 7 The spatial distribution of the content scores measured by MALDI IMS-MSI lipids is shown. In the control area outside the tissue sample (the square area in the lower left corner), the determined scores are significantly lower than those in the tissue itself. The dark areas in the sample correspond to denser tissue areas, which have a higher lipid signal ratio relative to the matrix background. DETAILED DESCRIPTION

[0041] While the invention has been shown and described with reference to several embodiments, those skilled in the art will recognize that various changes in form and details may be made without departing from the technical teachings of the invention as defined in the appended claims.

[0042] In the following, a method for evaluating spatially resolved ion mobility spectrometry-mass spectrometry data is described, which provides a score or quantitative value for each region, which indicates approximately how much percentage of the measured spectrum there originates from analytically interesting species of the sample. A high value of this quantitative value means that the proportion of molecules of interesting species ionized from the sample is high, so the information content is high and the signal quality is high. On the contrary, a low value can indicate that the measured spectrum is mainly a background signal, which can be traced back to the sample preparation method and / or the ionization method, for example.

[0043] Figure 2 is a schematic diagram of a possible ion mobility spectrometer-mass spectrometer to acquire spatially resolved collision cross section-mass signal histograms from a two-dimensional or flat tissue sample, see the setup in Part A. The setup and operation are described very briefly here:

[0044] A laser system (Bruker SmartBeam 3-D, top left) with various optical components is designed to bombard a flat sample on a sample holder with pulses. The sample holder can be scanned stepwise to obtain spatially resolved measurement signals from flat samples, e.g., wide two-dimensional samples such as flat tissue sections, or arrays of individually prepared samples, such as topically applied tissue homogenate preparations.

[0045] Once generated, the ions enter an ion mobility spectrometer (dual TIMS cell) which has an accumulation section and an analysis section. An inert gas flows through the two sections of the dual TIMS cell (from left to right in the figure). In the cell, the gas flow drives the ions against the opposing electric field, see detail B in the middle for an illustration of the principle. In the analysis section, the ions are separated at different positions on the axis according to their mobility.

[0046] The gradual reduction of the electric field strength in the analysis section of the dual TIMS cell allows the sequential release of ions separated according to mobility (detail B, scan). After the mobility analysis in the analysis section, the ions collected in the accumulation section at the same time will be transferred to the analysis section (detail B, pulse). The ions leaving the analysis section first pass through the ion transfer multipole and then enter the quadrupole mass filter. Here, ions can be selected for further analysis while other ions are removed. The ions are then transferred to the collision cell, where the selected ions are fragmented by accelerating injection into a neutral gas.

[0047] The ions and / or any fragment ions generated therefrom are temporarily stored in a collision cell before being introduced into the ion pulser of the time-of-flight analyzer by orthogonal injection in a time-coordinated manner. There they are accelerated perpendicularly to the injection direction onto the flight path of the reflectron time-of-flight analyzer. At the end of the flight path, a detector (not shown) receives the different ion packets with time resolution and therefore mass resolution and outputs them as time-of-flight transients, which can then be rescaled to mass (m) or mass / charge ratio (m / z).

[0048] As above, a so-called trapped ion mobility spectrometer (TIMS) coupled to a time-of-flight mass analyzer is described with reference to Figure 2. It should be understood that other embodiments of mobility spectrometers and mass analyzers can be used to generate collision cross section-mass signal histograms, such as drift tube ion mobility spectrometers, in which the neutral gas is at rest or flows in the opposite direction to the ion motion, and an electric field is used, and which is coupled to a cyclotron resonance cell, for example.

[0049] Description of the preferred method

[0050] The method described here is applicable to IMS data collected in conjunction with a MALDI MSI ion source (MALDI IMS-MSI). In a preferred embodiment, the method at least comprises the following steps:

[0051] 1. Calculation of a superimposed collision cross section-mass signal histogram which collates the information at each measurement point of the flat sample from a plurality (preferably all) of histograms.

[0052] 2. Determine the signal portion of interest (first choice of ion species), for example by regression analysis to determine the correlation between the collision cross section and the mass of the main signal portion of the complete measurement.

[0053] 3. A portion of the specified collision cross section-mass plane (σ-m plane or the plane of quantities derived therefrom) which contains the main and meaningful signal portion of the measurement ("signal channel") and, if applicable, a second portion of the specified collision cross section-mass plane which has a further signal portion of slightly less significance.

[0054] 4. By comparing the signal portions inside and outside the signal channel or comparing the signal portions within two previously defined (preferably separated) segments (e.g., covering different charge states (z) of molecules and / or molecular classes), a content score is calculated for each measurement point of a two-dimensional or flat sample, which can be used as a signal quality score.

[0055] 5. Evaluate the content score for subsequent analytical steps. In the following, these steps will be explained in more detail.

[0056] Collision cross section-mass signal histogram superposition

[0057] In a first step, a superimposed collision cross section-mass signal histogram can be formed, which combines the information from preferably all histograms for the individual measuring points. In order to shorten the calculation time, it is also possible to use selected representative histograms to form the superimposed collision cross section-mass signal histogram. For example, the superposition can be carried out by a simple summation, that is, the individual histograms are summed in terms of intensity (J) to form a total histogram. However, more complex superpositions are also conceivable, for example, so that in each of the included histograms, only a certain threshold value (J) of the intensity is taken into account. min ) above. The signal strengths of the individual histograms can also be modified by the transformation before forming the sum, e.g., depending on the purpose of the evaluation, more or less emphasis can be placed on parts with strong signals than on parts with weaker signals.

[0058] Regression analysis

[0059] It is assumed that the correlation between the collision cross section σ and the mass m (or the correlation of the corresponding derived quantity) for the meaningful signal part of the complete measurement (e.g. the main signal part) is approximately described by a power law with parameters C and k as described above. These parameters can be estimated based on the superimposed histogram by using an appropriate regression analysis. The superimposed histogram consists of a plurality of individual signals, each of which corresponds to a tuple (e.g. m i , σ i , J i ), where m i Indicates the corresponding mass (may also be (m / z) i , mass-to-charge ratio), σ i represents the collision cross section (and possibly also the derived mobility quantity Ki ), and J i Represents the measured signal strength. There are several methods for actual regression analysis, including the following in particular:

[0060] Simple logarithmic regression

[0061] By taking the logarithm of the above power law, a linear relationship can be obtained, which can be described by the model equation

[0062] log(σ i )=k log(m i )+log(C)+ε i .

[0063] Here, ε i represents the deviation from the model, assuming that the deviation is randomly distributed. Through simple linear regression, the constants C and k can be determined as follows:

[0064]

[0065] and

[0066]

[0067] in

[0068]

[0069] The above equation can be easily adapted to quantities derived from the collision cross section (σ) and mass (m).

[0070] Robust Logistic Regression

[0071] A known disadvantage of the simple logarithmic regression described above is the sensitivity to outliers in the measured data. Alternatively, a more robust regression method can be used to estimate the parameters k and log(C) of the linear model equation. In particular, two possible methods can be used, namely (i) lasso regression (Tibshirani, Robert (1996), "Regression Shrinkage and Selection via the lasso", Journal of the Royal Statistical Society, Series B (Methods), Wiley.58 (1): 267–88) and / or (ii) Theil-Sen estimator (see Sen, Pranab Kumar (1968), "Estimates of regression coefficient based on Kendall's tau", Journal of the American Statistical Association, 63 (324): 1379-1389).

[0072] Log-Radon Transform

[0073] Another possibility for regression analysis is to apply the Radon transform to the 2D histogram of the measured intensities in the logarithmic σ-m plane (see Radon, Johann (1917), “ die Bestimmune von Funktionen durch ihreIntegralwerte gewisser Mannigfaltigkeiten", Akademie der Wissenschaften zu Leipzig, Mathematics-Physics, Leipzig: Tübner (69): 262–277). The range of values ​​covered is subdivided into rectangular subsections of equal size. For each subsection, its associated measurement value ( logσ i ) Those intensities J that fall into this sub-segment i Perform the summation (see Figure 5 ). Apply the Radon transform to the resulting 2D histogram and find the position with the maximum intensity in it (see Figure 6 The inverse transformation of this position gives the desired straight line in the logarithmic σ-m plane (see Figure 5 ). The parameters of this straight line - slope k and y-intercept log(C) - ultimately describe the desired correlation between the collision cross section σ and the mass m according to a power law in the form of a curve that represents the contour of the main and meaningful signal part of the superimposed collision cross section-mass signal histogram. Here, the method can also be applied to quantities derived from the collision cross section (σ) and the mass (m).

[0074] Specify a channel or region of a signal portion

[0075] After determining the scaling factor C and the exponent k of the power law, the factor C can be set lo <C and C hi > C (lo stands for low; hi stands for high). When these values ​​are substituted for C in the power law, offset curves are obtained that specify the boundaries of the signal path (see Figure 5 ). By appropriately choosing C lo and C hi , these curves enclose the main signal part. By choosing C lo = 0, the signal channel can be extended downward and to the right until the boundary of the histogram, so that the channel is actually bounded only at the top. Alternatively, the signal channel can be extended downward and to the right until the boundary of the histogram, so that the channel is actually bounded only at the top.hi =∞(infinite) extends to the upper left. C lo and C hi The appropriate choice of depends on the molecular species being analyzed. For example, typical values ​​for lipids and glycans are C lo =0.90C…0.96C and C hi =1.04C…1.10C. In an alternative embodiment, a similar method can be used to determine a second part (second choice) of the collision cross section-mass plane, which should not contain all measurement signals outside the previously determined signal channel or range, but only certain ion species, especially those meaningless species, such as MALDI matrix clusters or similar background. Here, the method can also be applied to quantities derived from the collision cross section (σ) and the mass (m).

[0076] Calculation of content score ("signal quality score")

[0077] The content fraction is calculated for each individual measurement point with the aid of the corresponding collision cross section-mass signal histogram. The relevant tuple entries for a single measurement point or a single histogram are, as described above, for example, represented by (m i , σ i , J i ). S represents the amount of marker i, for which the corresponding measured value (m i , σ i ) is located in the signal channel or range. Then the content score G of the considered histogram can be calculated using the following formula

[0078]

[0079] Thus, in this example, the measurement signal from a signal channel or range is compared to all measurement signals (both within and outside the channel or range). This calculation method has proven to be particularly stable in terms of data processing. The second selection of ion species (e.g., meaningless signal parts) includes all measurement signals that are not within the preferred signal channel or range, which in turn determines the measurement signal of the first selected ion species in this embodiment.

[0080] However, in other embodiments, a calculation method is also possible that directly compares the first selection of ion species with the second selection, for example according to the following equation:

[0081]

[0082] This calculation method provides content scores that represent the actual content ratio of each individually selected ion species and are therefore easier to interpret.

[0083] Furthermore, it is conceivable and intended to define a second selection of ion species such that it does not include all ion species that are not assigned to the first selection of ion species. For example, a second signal channel or range can be determined for a specific second signal portion in the collision cross section-mass plane (or in the plane of the corresponding derived quantities). This approach is useful in such a measurement signal histogram, that is, it contains signal portions of several different substance classes, such as lipids, polysaccharides and peptides, which are candidates for meaningful signal portions in principle as biomolecules, as well as matrix clusters or other background ion species, which are essentially meaningless. The content fraction can be calculated as before by:

[0084] or

[0085] Where S a and S b respectively refer to the quantities of the labels i for which the corresponding measurement signal tuple entries of the individual measurement signal histograms (e.g., m i , σ i The first or second selection of ion species (or the amount derived therefrom) is assigned to each. Furthermore, the advantage of the second calculation method described above is that, for example, when the first and second selection of ion species contain different biomolecules from a two-dimensional or flat sample and their ratio is calculated, the content fractions can simply be reversed, thus producing an informative result.

[0086] Evaluate the content score (“Signal Quality Score”)

[0087] The content fraction can be assessed by visualizing the spatial distribution of the spatially resolved content fraction as a gray scale or pseudo color image (see Figure 7 ). Alternatively, the content scores can be used to control subsequent analysis of the data. For example, only those measurement points and histograms whose content scores are above a certain threshold, below a certain maximum value, or within a certain value range can be used for analysis. The content scores can also be used as weighting factors, so that those measurement data with higher content scores (and therefore presumably better quality) have a higher weighting than those with lower scores. For example, different weightings can be used for adaptive noise suppression in subsequent data analysis.

[0088] In addition to the embodiments described by way of example, other embodiments of the present invention may be conceivable. With the knowledge of this disclosure, those skilled in the art can easily design other advantageous embodiments, which are also covered by the protection scope of the claims, including any equivalent forms as appropriate.

Claims

1. A method for processing spatially resolved ion spectroscopic measurement signal data recorded on a two-dimensional sample, the method comprising: - providing measurement signal data with a plurality of measurement signal histograms, wherein the measurement signal histograms are assigned to a limited area (A) of the two-dimensional sample by means of two position coordinates (x, y) fin，x，y ), the area is smaller than the total area of ​​the two-dimensional sample (A total ), the histogram further comprising measurement signal tuples having an intensity dimension (J) or a quantity derived therefrom, a mass dimension (m) or a quantity derived therefrom and a collision cross section dimension (σ) or a quantity derived therefrom; - assigning a first selection of ion species comprising biomolecules and a second selection of ion species for the two-dimensional sample, the presence of these selected ion species in the measurement signal histogram being detectable and distinguishable using the collision cross section dimension (σ) or a quantity derived therefrom, wherein the first selection of ion species is assigned by summing several measurement signal histograms into a superimposed measurement signal histogram and determining a meaningful portion of measurement signal tuple entries in the superimposed measurement signal histogram by regression analysis; - Determine the limited area (A fin，x，y ) in the measured signal histogram of the first selected ion species and the spatially resolved content of the second selected ion species, and calculating these different contents to form a spatially resolved content fraction (G x，y );and -Using spatially resolved content fraction (G x，y ) to mark the two-dimensional sample. 2 . The method according to claim 1 , wherein the two-dimensional sample is treated with a matrix substance for matrix-assisted laser desorption.

3. The method of claim 1, wherein the first selection of ion species includes species of high analytical significance and the second selection of ion species includes species of low analytical significance.

4. The method of claim 1, wherein the first selection of ionic species comprises proteins, peptides, glycans and / or lipids.

5. The method of claim 1, wherein the second selection of ion species comprises charged atoms or molecules and / or clusters thereof produced by a sample preparation method and / or an ionization method.

6. The method of claim 1, wherein after marking, an image of the two-dimensional sample is presented to a user, wherein: Each limited area (A fin，x，y ) with a specified content fraction (G x，y ) are coded visually.

7. The method according to claim 1, wherein the predetermined value range of the content fraction is evaluated as an ion spectrum measurement signal from outside the two-dimensional sample.

8. The method according to claim 1, wherein the subsequent evaluation (i) takes into account only the content fraction (G x，y ) is within a limited area within a predetermined value range (A fin，x，y ) and / or (ii) using the content fraction (G x，y ) as a weighting factor. 9 . The method according to claim 1 , wherein all measurement signal histograms or a representative selection of measurement signal histograms are added to form the superimposed measurement signal histogram.

10. The method of claim 1, wherein in at least one dimension of the measurement signal histogram, a meaningful portion of the measurement signal tuple entries are used to distinguish between the first selection and the second selection of ion species.

11. The method of claim 1, wherein the regression analysis comprises lasso regression or Theil-Sen estimator.

12. The method of claim 1, wherein the regression analysis comprises logarithmic regression or logarithmic Radon transformation.

13. The method of claim 1, wherein the regression analysis is based on the equation σ(m)≈Cm k to explore the correlation between the collision cross section dimension (σ) and the mass dimension (m) of a molecule, where C is a molecule-dependent scaling factor and k is a molecule-dependent exponent. 14 . The method as claimed in claim 1 , wherein the superimposed measurement signal histogram is calculated by a position-independent addition of a plurality of measurement signal histograms.

15. A method according to claim 14, wherein the position-independent summation only takes into account such measurement signal histograms, in which the measurement signal tuple entries of at least one dimension (i) exceed a predetermined threshold, (ii) are below a predetermined threshold, or (iii) are within a predetermined value range. 16 . The method according to claim 14 , wherein at least one measurement signal tuple entry of the respective measurement signal histogram is transformed before the summing up, such that measurement signal tuple entries of a first predetermined range are weighted disproportionately relative to a second predetermined range.

17. The method according to claim 1, wherein the spatially resolved content fraction (G x，y ) is calculated as follows Where S a and S b are the quantities for the labels i for which the corresponding measurement signal tuple entries (m i , σ i ) are assigned to the first and second choices of ion species, respectively.

18. A method for acquiring and processing ion spectrum measurement signal data, wherein (i) the ion spectrum measurement signal data is acquired by means of an ion mobility spectrometer-mass spectrometer, and (ii) the method according to any one of claims 1 to 17 is performed.

19. Ion mobility spectrometer-mass spectrometer having a computing and / or control unit, designed and configured to carry out the method according to any one of claims 1 to 17.

Images