Retention time correction method and computer readable storage medium

Abstract

The application provides a retention time correction method and a computer readable storage medium, and relates to the technical field of metabolomics. The retention time correction method provided by the application comprises the following steps: determining endogenous dicarboxylic acids and the retention times of the endogenous dicarboxylic acids under actual test conditions according to the data of dicarboxylic acid standard product series and metabolomics data of a biological matrix to be measured, wherein the metabolomics data are data obtained by analyzing the biological matrix to be measured under actual test conditions by using negative ion mode ultra-high performance liquid chromatography-mass spectrometry; taking the retention times of two endogenous dicarboxylic acids with adjacent carbon atom numbers as references, calculating the retention indices of each metabolite to be measured with a retention time between the two endogenous dicarboxylic acids; and correcting the retention time of each metabolite to be measured under the actual test conditions to the retention time under standard test conditions according to the retention indices of each metabolite to be measured. In this way, the retention time deviation of metabolites caused by differences in test conditions can be effectively reduced.

Description

Technical Field

[0001] This application relates to the field of metabolomics technology, and in particular to a retention time correction method and a computer-readable storage medium for liquid chromatography-mass spectrometry analysis. Background Technology

[0002] Ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) technology has been widely used for the identification and quantification of metabolites in biological samples. Reliable metabolite identification requires determining the retention time of the analyte in liquid chromatography (HPLC). t R The mass spectrometry data are matched with their corresponding standards. In particular, retention time is a key orthogonal data point that provides additional information about the interactions between the analyte, stationary phase, and elution solvent, complementing the mass spectrometry data and significantly improving the accuracy of identification.

[0003] However, many factors, such as mobile phase composition, flow rate, elution gradient, and column type, can cause shifts in the retention time of metabolites, making it impossible to match and compare metabolomics data with metabolite standard databases from different laboratories. This problem directly limits the accurate identification of known small molecule metabolites using liquid chromatography-mass spectrometry (LC-MS), thus restricting its application in metabolomics.

[0004] Retention Index ( RI The retention index correction method establishes a relative retention scale using a reference, which can effectively correct for retention time drift and enable cross-platform data comparison. It has been widely used in LC-MS metabolomics research. Existing retention index correction methods mostly use exogenous additives as references, such as adding chemically labeled mixtures of saturated fatty acids and fatty amines for reversed-phase chromatography (RPLC) and hydrophilic chromatography (HILIC) (Anal. Chem. 2018, 90(14): 8412-8420; Anal. Chem. 2019, 91(9): 6057-6063). Another method uses N-alkylpyridine-3-sulfonates with different alkyl chain lengths (C1 to C20) as references and establishes standardized retention indices for 96 fungal toxins and their metabolites under different liquid chromatography conditions (J. Chromatogr. A 2021, 1639: 461901). All of the above studies have shown that the retention time corrected by the retention index has higher reproducibility than the original retention time. However, adding additional exogenous compounds inevitably increases the complexity of the sample, alters the sample matrix, requires additional sample preparation and mass spectrometry acquisition resources, and interferes with the detection sensitivity and stability of the target metabolite. Furthermore, the above correction strategies are highly dependent on the retention time and mass spectrometry data of these exogenous references in the metabolomics data, limiting their ability to retrospectively correct historical data.

[0005] Existing patent CN112255361B has screened out a series of linear acylcarnitines (C0-C26) that can be detected in positive ion mode from biological samples, and constructed a retention time correction method based on the quantitative relationship between the retention time of acylcarnitine and the number of carbon atoms, so as to effectively correct the retention time deviation caused by multiple factors.

[0006] However, the aforementioned acylcarnitine-based correction methods strictly rely on positive ion mode mass spectrometry detection. Acylcarnitine, the reference compound, exhibits extremely low or even no signal response in negative ion mode, making it unsuitable for direct application to metabolomics data acquired in negative ion mode. Conversely, usable endogenous reference compounds in negative ion mode must simultaneously meet two stringent conditions: "stable and high response in negative ion mode" and "retention time covering the metabolite peak range." Since dicarboxylic acid series compounds do not produce peaks in positive ion mode, and acylcarnitine does not produce peaks in negative ion mode, they are fundamentally isolated from each other based on the physicochemical basis of mass spectrometry detection. Therefore, for metabolomics data in negative ion mode, a usable retention time correction scheme cannot be obtained simply by replacing acylcarnitine with other endogenous substances. Thus, developing a retention time correction method specifically for ultra-high performance liquid chromatography-mass spectrometry analysis in negative ion mode remains a pressing technical challenge. Summary of the Invention

[0007] This application provides a retention time correction method and a computer-readable storage medium specifically for ultra-high performance liquid chromatography-mass spectrometry analysis in negative ion mode. It aims to solve the problem of metabolite retention time correction caused by the lack of suitable endogenous references in negative ion mode, and can effectively reduce the retention time deviation of metabolites caused by differences in test conditions.

[0008] In a first aspect, this application provides a retention time correction method, including steps S10 to S30: S10: Based on the data of a series of dicarboxylic acid standards with sequentially increasing carbon atom numbers and the metabolomics data of the target biological matrix, determine the endogenous dicarboxylic acid series corresponding to the dicarboxylic acid standards in the target biological matrix, and determine the retention time of each endogenous dicarboxylic acid under actual test conditions. The data for each dicarboxylic acid standard are obtained by ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS / MS) under standard test conditions, including its standard retention time, precise mass-to-charge ratio, and fragment ion structure. The metabolomics data are obtained by ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS / MS) under actual test conditions, including the retention time, precise mass-to-charge ratio, and fragment ion structure of endogenous dicarboxylic acids and each metabolite to be tested. S20: Using the retention times of two endogenous dicarboxylic acids with adjacent carbon atom numbers as a reference, calculate the retention index of each analyte metabolite whose retention time falls between the two; and S30: Based on the retention index of each analyte, correct the retention time of each analyte under actual test conditions to the retention time under standard test conditions.

[0009] Thus, by selecting endogenous dicarboxylic acids as references for liquid chromatography-mass spectrometry (LC-MS), and based on the data of dicarboxylic acid standards and the metabolomics data of the target biological matrix, the retention times of each endogenous dicarboxylic acid under actual test conditions are obtained. Based on the retention times of each endogenous dicarboxylic acid under actual test conditions, the retention index of each analyte metabolite is calculated, and the retention times under actual test conditions are corrected to the retention times under standard test conditions based on the retention index. This method can effectively reduce the retention time deviation of metabolites caused by differences in test conditions.

[0010] In another illustrative embodiment of the retention time correction method, the dicarboxylic acid standard series is a series of saturated straight-chain alkyl dicarboxylic acids with 2 to 26 carbon atoms, increasing sequentially. The retention time range formed by these dicarboxylic acid standards can cover the retention time range of endogenous metabolites under ultra-high performance liquid chromatography.

[0011] Furthermore, step S10 in the retention time correction method specifically includes S11 to S16: S11: Based on data from a series of dicarboxylic acid standards, spectral matching and peak extraction were performed on metabolomics data to determine the retention times of each endogenous dicarboxylic acid under actual testing conditions. S12: Determine whether the identified endogenous dicarboxylic acids correspond one-to-one with the dicarboxylic acid standard series in terms of the number of carbon atoms. If so, proceed to S13: Output all identified endogenous dicarboxylic acids and their retention times under actual test conditions. If not, proceed to S14: Construct a cubic regression equation with the number of carbon atoms in the dicarboxylic acid standard as the independent variable and the standard retention time of the dicarboxylic acid standard under standard test conditions as the dependent variable. S15: Using the identified endogenous dicarboxylic acids and their retention times under actual testing conditions as data points, the regression equation is fitted with coefficients to obtain a predictive equation. The predictive equation is a cubic equation with the number of carbon atoms in the endogenous dicarboxylic acid as the independent variable and the retention time of the endogenous dicarboxylic acid under actual testing conditions as the dependent variable. S16: Based on the prediction equation, predict the retention time of undetected endogenous dicarboxylic acids in the endogenous dicarboxylic acid series under actual test conditions, and output all endogenous dicarboxylic acids and their retention times under actual test conditions.

[0012] This facilitates the accurate and comprehensive acquisition of the retention times of each endogenous dicarboxylic acid under the test conditions.

[0013] Furthermore, the regression equation in step S14 is expressed as a polynomial as shown in equation (1): t R sd = 0.0008 L c 3 – 0.0457 L c 2 +1.1828 L c – Equation (1) 2.4883 in, t R sd The standard retention time of dicarboxylic acid standards under standard test conditions; L c The number of carbon atoms in the dicarboxylic acid standard.

[0014] In this way, the quantitative relationship between the number of carbon atoms in dicarboxylic acid standards and the standard retention time can be accurately fitted.

[0015] Furthermore, the prediction equation for step S15 is expressed as a polynomial as shown in equation (2): t R = 0.0007 L c 3 – 0.0426 L c 2 +1.1391 L c – Formula 2.2965 (2), in, t R The retention time of endogenous dicarboxylic acids under actual test conditions; L cThe number of carbon atoms in an endogenous dicarboxylic acid.

[0016] In this way, the quantitative relationship between the number of carbon atoms and retention time of endogenous dicarboxylic acids can be accurately fitted, which is beneficial for predicting the retention time of undetected endogenous dicarboxylic acids.

[0017] Furthermore, in step S20 of the retention time correction method, the retention index of each analyte metabolite is calculated using the formula shown in equation (3): Equation (3), in, Under actual test conditions, these represent the retention times of the analyte metabolite i and the retention times of two endogenous dicarboxylic acids with adjacent carbon atoms m and n, respectively. ; These are the retention indices of the analyte metabolite i and the retention indices of two endogenous dicarboxylic acids, m and n respectively, which are adjacent to the analyte metabolite i in the efflux sequence, under actual test conditions.

[0018] This facilitates the accurate calculation of the retention index of each metabolite to be tested.

[0019] Furthermore, the retention index of endogenous dicarboxylic acids is set to 100 times the number of carbon atoms in the endogenous dicarboxylic acid. This makes it easier to calculate the retention index of each endogenous dicarboxylic acid.

[0020] Furthermore, in step S30 of the retention time correction method, the retention time of each analyte metabolite is corrected using the formula shown in equation (4): Equation (4) in, The corrected retention time of the analyte metabolite i under standard test conditions. The standard retention times are those of two dicarboxylic acid standards, m and n, which are adjacent to the elution sequence of the metabolite i. .

[0021] This facilitates the accurate correction of the retention time of the metabolite under actual test conditions to its retention time under standard test conditions.

[0022] Furthermore, the chromatographic parameters for both standard and actual test conditions include the elution gradient, mobile phase flow rate, column type, and column temperature. This facilitates the analysis of the effectiveness of the retention time correction method.

[0023] Secondly, this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the aforementioned retention time correction method. This method can effectively reduce the retention time deviation of metabolites caused by differences in test conditions. Attached Figure Description

[0024] The following figures are for illustrative purposes only and do not limit the scope of the invention.

[0025] Figure 1 A schematic flowchart illustrating one implementation of the retention time correction method.

[0026] Figure 2 for Figure 1 The flowchart shown is a partial flowchart of the retention time correction method.

[0027] Figure 3 This is used to illustrate the secondary mass spectrum of hexadecanoic acid and its mass spectrometric fragmentation behavior.

[0028] Figure 4 Used to illustrate regression and prediction equations.

[0029] Figure 5 The diagram shows the extraction chromatograms of various biological matrix samples and the elution sequence of endogenous dicarboxylic acids and analyte metabolites.

[0030] Figure 6 This is used to illustrate the changes in the retention time deviation of each analyte metabolite before and after correction under different actual test conditions.

[0031] Figure 7 This is used to illustrate the effect of the retention time correction method on the retention time deviation of the analyte obtained under the actual test conditions of M3.

[0032] Figure 8 This is used to illustrate the effectiveness of the retention time correction method in correcting retention time deviations of analyte metabolites across different batches and platforms. Detailed Implementation

[0033] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0034] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein in the specification of the application is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims and drawings of this application are intended to cover non-exclusive inclusion.

[0035] The term "embodiment" as used herein means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of the phrase "embodiment" in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0036] In this article, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists, A and B exist simultaneously, or B exists. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0037] Furthermore, the terms "first," "second," etc., in the specification and claims of this application or in the aforementioned drawings are used to distinguish different objects rather than to describe a specific order, and may explicitly or implicitly include one or more of the features.

[0038] In the description of this application, unless otherwise stated, "multiple" means two or more (including two), and similarly, "multiple groups" means two or more (including two groups).

[0039] It should be understood that the phrase "one embodiment" or "an embodiment" throughout the specification means that a specific feature, structure, or characteristic related to the embodiment is included in at least one embodiment of this application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. Furthermore, these specific features, structures, or characteristics can be combined in any suitable manner in one or more embodiments.

[0040] Based on the above, this application provides a retention time correction method, which will be described in detail below with reference to the accompanying drawings.

[0041] Figure 1 This is a schematic flowchart illustrating one embodiment of the retention time correction method. In this illustrative embodiment, the retention time correction method includes steps S10 to S30.

[0042] S10: Based on the data of the series of dicarboxylic acid standards with sequentially increasing carbon atom numbers and the metabolomics data of the target biological matrix, determine the endogenous dicarboxylic acid series in the target biological matrix corresponding to the dicarboxylic acid standards series, and determine the retention time of each endogenous dicarboxylic acid under actual test conditions.

[0043] Retention time ( t R Retention time refers to the time required for a separated component to reach its maximum concentration within the chromatographic column from the start of injection, usually measured in minutes. Retention time directly reflects the interaction force between the metabolite and the stationary phase (packing material) in the column. The stronger the interaction, the slower the column passage, and the longer the retention time. It is the most fundamental raw data in chromatographic separation.

[0044] The data for each dicarboxylic acid standard series are obtained by ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS / MS) analysis under standard test conditions in negative ion mode. These data include the standard retention time, precise mass-to-charge ratio, and fragment ion structure of the dicarboxylic acid. The dicarboxylic acid standards include, for example, 25 saturated straight-chain alkyl dicarboxylic acids with sequentially increasing carbon atom numbers: oxalic acid, malonic acid, succinic acid, glutaric acid… up to hexadecanoic acid.

[0045] In UHPLC-MS metabolomics analysis, selecting endogenous dicarboxylic acids as references offers significant scientific advantages. First, dicarboxylic acids (such as oxalic acid, succinic acid, and adipic acid) are typical normal metabolic products in vivo, belonging to endogenous compounds and widely involved in core physiological pathways such as the tricarboxylic acid cycle, fatty acid β-oxidation, and amino acid metabolism. This attribute determines that they naturally exist in various biological matrices (plasma, urine, tissues), serving as natural "internal benchmarks" without exogenous addition. Second, the retention time range of these dicarboxylic acids can cover the retention time range of endogenous metabolites under UHPLC. Furthermore, since the reference and analyte are both endogenous matrix components, their recovery behavior during pretreatment and their ion inhibition / enhancement effects during electrospray ionization are highly consistent with the low-abundance target metabolites, providing more accurate response correction. Compared to external standard methods or adding synthetic isotope internal standards, using endogenous dicarboxylic acids as references avoids impurity peaks introduced by insufficient reagent purity, and there is no need to worry about the stability of exogenous standards in complex biological samples.

[0046] Metabolomics data are obtained by analyzing the target biological matrix under actual testing conditions using ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS / MS) in negative ion mode. Metabolomics data typically include characteristic peak numbers of endogenous dicarboxylic acids and each analyte metabolite, retention times under actual testing conditions, precise mass-to-charge ratio of the parent ion, fragment ion structure information, peak height, and peak area. It records the relative quantitative trajectory of metabolites, locks in their precise mass numbers, retention time behavior, and fragmentation patterns, and serves as a data carrier for elucidating unknown metabolites.

[0047] Standard test conditions refer to a specific set of chromatographic conditions used to separate and identify known metabolites and their standards, and to determine their retention times. Chromatographic parameters for standard and actual test conditions include, for example, elution gradient, mobile phase flow rate, column type, and column temperature. Specifically, in one illustrative embodiment, the chromatographic parameters for standard and actual test conditions are shown in Tables 1 and 2, for example. Wherein, M... O The standard test conditions used for ultra-high performance liquid chromatography-mass spectrometry analysis in negative ion mode are M1 to M2. 11 These represent the different actual test conditions used in ultra-high performance liquid chromatography-mass spectrometry analysis under negative ion mode.

[0048] Table 1. Chromatographic parameters under standard test conditions and different actual test conditions.

[0049] The four chromatographic columns (injection volume of 1 μL) in Table 1 are as follows: C1: Waters ACQUITY UPLC HSS T3 column (2.1*100mm, 1.8μm), C2: Agilent ZORBAX RRHD Eclipse Plus C18 column (2.1*100mm, 1.8μm), C3: Waters ACQUITY UPLC BEH C18 column (2.1*100mm, 1.7μm), and C4: Waters ACQUITY UPLC BEH C18 column (2.1*50mm, 1.7μm).

[0050] Table 2 Five Chromatographic Gradients

[0051] Where B represents acetonitrile, and B% indicates the volume percentage of acetonitrile in the elution solvent.

[0052] By comparing the retention time, precise mass-to-charge ratio, and fragment ion structure of endogenous dicarboxylic acids in metabolomics data under actual test conditions with the standard retention time, precise mass-to-charge ratio, and fragment ion structure of dicarboxylic acid standards under standard test conditions, endogenous dicarboxylic acids in biological matrices can be identified and their retention time under actual test conditions can be obtained.

[0053] S20: Using the retention times of two endogenous dicarboxylic acids with adjacent carbon atom numbers as a reference, calculate the retention index of each metabolite whose retention time falls between the two.

[0054] The retention index (RI) is a dimensionless value obtained by mathematically standardizing and relativizing absolute retention times. It is a "standardized fraction" or "relative position coordinate" characterizing the retention time of metabolites on a chromatographic column. Those skilled in the art often combine the retention index, precise mass-to-charge ratio, and fragment ion structure to identify substances.

[0055] S30: Based on the retention index of each analyte, correct the retention time of each analyte under actual test conditions to the retention time under standard test conditions.

[0056] By selecting endogenous dicarboxylic acids as references for liquid chromatography-mass spectrometry (LC-MS), and based on data from a series of dicarboxylic acid standards and metabolomics data of the target biological matrix, the retention times of each endogenous dicarboxylic acid under actual test conditions were determined. Based on the retention times of each endogenous dicarboxylic acid under actual test conditions, the retention index of each analyte metabolite was calculated. The retention times under actual test conditions were then corrected to those under standard test conditions based on the retention index. This method effectively reduces the retention time deviation of metabolites caused by differences in test conditions.

[0057] Figure 2 for Figure 1 The flowchart shown is a partial representation of the retention time correction method. See also... Figure 2 In the illustrative embodiment, step S10 specifically includes S11 to S16.

[0058] S11: Based on the data of the dicarboxylic acid standard series, spectral matching and peak extraction were performed on the metabolomics data to determine the retention time of each endogenous dicarboxylic acid under actual test conditions.

[0059] Analysis using ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) in negative ion scanning mode revealed that dicarboxylic acids readily lose a hydrogen ion from their carboxyl groups in negative ion mode, forming a precursor ion [MH]. - In addition, the parent ion [MH] -In the collision cell, the ion is broken up to produce daughter ions. These daughter ions may be, for example, fragment ions that have lost one molecule of water, one molecule of carbon dioxide, and / or one molecule of formic acid. Within a defined space of retention time and parent ion mass-to-charge ratio, ion swarm signals exhibiting "highly synchronized retention times and integer mass differences in mass-to-charge ratio" are classified as primary and secondary mass spectrometry signals of the same dicarboxylic acid. The primary and secondary mass spectrometry signals are then compared with data from dicarboxylic acid standards to determine the name of the dicarboxylic acid and its retention time under actual testing conditions.

[0060] The following uses hexadecanoic acid as an example to illustrate spectral matching and peak extraction for the detection of hexadecanoic acid. Figure 3 This is used to illustrate the secondary mass spectrum of hexadecanoic acid and its mass spectrometric fragmentation pattern. Table 3 shows the saturated straight-chain alkyl dicarboxylic acid standards and their detection in different biological matrices. The dicarboxylic acid standards are saturated straight-chain alkyl dicarboxylic acids with the number of carbon atoms continuously increasing from 2 to 26.

[0061] Table 3. Standards of saturated straight-chain alkyl dicarboxylic acids and their detection in different biological matrices.

[0062] Wherein, “√” indicates that the compound is strongly detected in the biological matrix, and “Δ” indicates that the compound is weakly detected in the biological matrix.

[0063] Combined with Table 3 and Figure 3 The parent ion [MH] formed from hexadecanoic acid - Its precise mass-to-charge ratio is 285.2071, and its structure is as follows: Figure 3 As shown in b, its spectrum can be found in [reference needed]. Figure 3 The mass spectrum peak of a at 285.2087 (within an error range of, for example, 2 mDa).

[0064] See Figure 3 b, the parent ion [MH] - The daughter ions produced by the breakup in the collision cell, for example, those that have lost one molecule of water, one molecule of carbon dioxide, and one molecule of formic acid, have mass-to-charge ratios of 241.2173, 267.1966, and 239.2017, respectively. These daughter ions can further lose one molecule of water or one molecule of carbon dioxide to produce new daughter ions with mass-to-charge ratios of 223.2067 and 221.1911, respectively. Figure 3 The value of a also clearly shows the peaks for mass-to-charge ratios of 223.2082 and 267.1980.

[0065] Within a defined space for hexadecanoic acid retention time (7.91 min, with an error of, for example, 0.1 min) and mass-to-charge ratio (285.2087, with an error of, for example, 2 mDa), the primary and secondary mass spectrometry signals of hexadecanoic acid were identified, for example... Figure 3 The peaks with mass-to-charge ratios of 223.2082 and 267.1980 shown in diagram a were compared with the spectra in the data of dicarboxylic acid standards. The compound was identified as hexadecanoic acid, and its corresponding retention time under actual test conditions was obtained from metabolomics data.

[0066] Similarly, those skilled in the art can detect other endogenous dicarboxylic acids and obtain their corresponding retention times using the methods described above.

[0067] S12: Determine whether the identified endogenous dicarboxylic acids correspond one-to-one with the dicarboxylic acid standard series in terms of the number of carbon atoms.

[0068] If so, proceed to S13: Output all identified endogenous dicarboxylic acids and their retention times under actual test conditions.

[0069] If not, proceed to S14: Construct a cubic regression equation with the number of carbon atoms in the dicarboxylic acid standard as the independent variable and the standard retention time of the dicarboxylic acid standard under standard test conditions as the dependent variable. Figure 4 Used to illustrate regression and prediction equations. See also Figure 4 , Figure 4 The 'a' in the figure illustrates an example of a regression equation. In the illustrative embodiment, the regression equation is expressed as a polynomial, for example, by equation (1), with a coefficient of determination R² = 0.9996: t R sd = 0.0008 Lc 3 -0.0457 Lc 2 + 1.1828 Lc –2.4883 (1) in, t R sd The standard retention time is the standard retention time of dicarboxylic acid standards under standard test conditions. t R sd These are the retention times of dicarboxylic acid standards obtained under standard test conditions. Table 3 shows the standard retention times of 25 dicarboxylic acids. Lc Table 3 shows the number of carbon atoms in 25 dicarboxylic acid standards.

[0070] Depend on Figure 4 From 'a', we can know the number of carbon atoms in a dicarboxylic acid. Lc The standard retention time of the dicarboxylic acid t R sd There is a good quantitative relationship between them, except for the retention time deviation of succinic acid ( Δt R Except for 0.16 min, the retention time deviations of other dicarboxylic acids were ( Δt R The values ​​are all less than 0.10 min, indicating that the regression equation can accurately fit the number of carbon atoms in dicarboxylic acids. Lc The standard retention time of the dicarboxylic acid t R sd The quantitative relationship between them. The retention time deviation here ( Δt R The difference between the retention time obtained by fitting the regression equation and the standard retention time is denoted as .

[0071] S15: Using the identified endogenous dicarboxylic acids and their retention times under actual testing conditions as data points, the regression equation is fitted with coefficients to obtain a predictive equation. The predictive equation is a cubic equation with the number of carbon atoms in the endogenous dicarboxylic acid as the independent variable and the retention time of the endogenous dicarboxylic acid under actual testing conditions as the dependent variable. Here, coefficient fitting refers to using the identified endogenous dicarboxylic acids and their retention times under actual testing conditions as data points, without changing the structure of the regression equation, only changing the coefficients, to fit the regression equation to obtain the predictive equation. That is, the regression equation and the predictive equation differ only in their coefficients.

[0072] As shown in Table 3, a few endogenous dicarboxylic acids in the biological matrix have weak detection peak intensities. As a result, the retention time of these endogenous dicarboxylic acids cannot be obtained from metabolomics data. Therefore, it is necessary to construct a prediction equation based on regression equation to predict the retention time of the above-mentioned endogenous dicarboxylic acids with weak detection peak intensities under actual test conditions.

[0073] Figure 4 The 'b' in the diagram schematically illustrates an example of a prediction equation. See also... Figure 4 In the illustrative embodiment, the prediction equation for b is expressed as a polynomial, for example, by equation (2), with a determination coefficient R² = 0.9998: t R = 0.0007 Lc 3 –0.0426 Lc 2+1.1391 Lc –2.2965 (2) in, t R The retention time of endogenous dicarboxylic acids under actual test conditions is obtained through step S11; Lc The number of carbon atoms in an endogenous dicarboxylic acid.

[0074] If succinic acid, icosanoic acid, and hexadecanoic acid are not detected in the biological matrix, based on the fact that the regression equation is a cubic equation, the coefficients of the remaining 22 dicarboxylic acids under actual testing conditions can be fitted to the regression equation to obtain a predictive equation. This predictive equation can be used to predict the retention times of succinic acid, icosanoic acid, and hexadecanoic acid under actual testing conditions. This allows for an accurate fitting of the quantitative relationship between the number of carbon atoms in dicarboxylic acids and their retention time, which is beneficial for predicting the retention times of undetected endogenous dicarboxylic acids under actual testing conditions.

[0075] S16: Based on the prediction equation, predict the retention time of undetected endogenous dicarboxylic acids in the endogenous dicarboxylic acid series under actual test conditions, and output all endogenous dicarboxylic acids and their retention times under actual test conditions.

[0076] See Figure 4 b, based on the prediction equation, can effectively predict the retention times of succinic acid, icosanoic acid, and hexadecanoic acid, and the retention time deviation of icosanoic acid and hexadecanoic acid is ( Δt R The retention time deviation of succinic acid is less than 0.12 min. Δt R Approximately 0.22 min. Therefore, based on the prediction equation, the retention time of undetectable endogenous dicarboxylic acids in different biological matrices under actual test conditions can be predicted with high precision. This facilitates the accurate and comprehensive acquisition of the retention time of each endogenous dicarboxylic acid under actual test conditions.

[0077] In an illustrative embodiment, in step S20, the retention index of each metabolite to be tested is calculated, for example, by the formula shown in equation (3): Equation (3), in, Under actual test conditions, these represent the retention times of the analyte metabolite i and the retention times of two endogenous dicarboxylic acids with adjacent carbon atoms m and n, respectively. ; These are the retention indices of the analyte metabolite i under actual testing conditions, and the retention indices of two endogenous dicarboxylic acids with adjacent carbon atoms m and n, respectively, to analyte i in the efflux sequence. For example, the retention index of an endogenous dicarboxylic acid is set to 100 times the number of carbon atoms of its constituent carbon atoms; for instance, the retention index of oxalic acid is 100 times the number of carbon atoms (200). This facilitates the accurate calculation of the retention index of each analyte metabolite.

[0078] In an illustrative embodiment, in step S30, the retention time of each analyte metabolite is corrected using the formula shown in equation (4): Equation (4) in, The corrected retention time of the analyte metabolite i under standard test conditions. The standard retention times are those of two dicarboxylic acid standards, m and n, which are adjacent to the elution sequence of the metabolite i. This facilitates the accurate correction of the retention time of the analyte under actual test conditions to the retention time under standard test conditions.

[0079] The present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the retention time correction method described above. This method can effectively reduce retention time deviations of metabolites caused by elution gradient, mobile phase flow rate, column type, and column temperature.

[0080] The following examples are used to illustrate the present invention, but are not intended to limit the scope of the invention.

[0081] Example Experimental instruments and materials 1. Instruments The ACQUITY® UHPLC liquid chromatography, Shimazu Exion LC20AD liquid chromatography, Vanquish Flex UHPLC liquid chromatography, Xevo G2-XS QTOF MS mass spectrometry, SCIEX Zeno TOF 7600 mass spectrometry and Orbitrap Exploris 240 mass spectrometry are used to form a UHPLC-MS coupled analytical system.

[0082] SCIEX OS (V2.1.6) and Masslynx™ (v4.1 SCN 966, Waters) are used for mass spectrometry data acquisition and processing.

[0083] 2. Reagents and medicines Formic acid and acetonitrile of mass spectrometry grade were purchased from Thermo Fisher Scientific (China) Co., Ltd. Ultrapure water used for sample processing and as the chromatographic mobile phase was obtained from Millimeter / Large Water Supply Co., Ltd. Q (Merck Millipore GmbH, Germany). Analytical purity standards were purchased from McLean Biochemical Technology Co., Ltd. (please specify the Chinese company name), Sanhua Chemical Co., Ltd. (please specify the Chinese company name), or Aladdin Chemical Reagent Co., Ltd. (Shanghai).

[0084] 3. Ultra-high performance liquid chromatography-mass spectrometry analysis All biological samples used in the following experiments were approved by the Ethics Committee of Fudan University. Human plasma was obtained from healthy adult volunteers in China who had signed informed consent forms recruited through the Human Phenome Project. Once obtained using standard clinical sampling methods, samples were immediately flash-frozen in liquid nitrogen and stored at -80°C for analysis. Animal tissue samples included rabbit liver tissue, human A549 cells, human serum, and human plasma, collected in accordance with the National Guidelines for Laboratory Animal Welfare (Ministry of Science and Technology, 2006).

[0085] For rabbit liver tissue or human A549 cells, weigh approximately 50 mg and add 600 μL of pre-cooled 80% methanol aqueous solution, homogenize for 6 min. Then, sonicate the sample in an ice bath for 10 min (45 kHz, 200 W), followed by centrifugation (14000 g, 10 min, 4 °C) and collect the supernatant. Repeat this process twice, combine the supernatants from all three cycles, and dry under nitrogen at room temperature. Finally, reconstitute the extract with 200 μL of 50% methanol aqueous solution, centrifuge again (14000 g, 10 min, 4 °C), and collect the supernatant for ultra-high performance liquid chromatography-mass spectrometry analysis.

[0086] For human serum or human plasma, take 50 μL of sample, add 200 μL of methanol, centrifuge (14000g, 10 min, 4℃), and take 200 μL of supernatant for analysis by ultra-high performance liquid chromatography-mass spectrometry.

[0087] For standard samples, use a mixed aqueous solution of saturated straight-chain dicarboxylic acid series standards as shown in Table 3.

[0088] Metabolomics data were acquired using four UHPLC-QTOFMS systems and four columns with different chemical properties. The four UHPLC-QTOFMS systems were as follows: the first and second systems both consisted of an ACQUITY® UHPLC system coupled with a Xevo G2-XS QTOFMS system; the third system consisted of a Shimazu Exion LC20AD system coupled with a SCIEX Zeno TOF 7600 system; and the fourth system consisted of a Vanquish Flex UHPLC system coupled with an Orbitrap Exploris 240 system. The four columns for each system are configured as shown in Table 1.

[0089] Ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS / MS) analysis was performed using water containing 0.1% formic acid and acetonitrile as the elution solvent, employing 12 elution methods M0 to M12. 11 As shown in Tables 1 and 2. Where M0 represents the standard test conditions, and M1 to M... 11 Eleven actual testing conditions were used. Metabolomics data were analyzed in positive and negative ion modes using MS. E The acquisition mode was configured with the following parameters: mass range of 50 m / z to 1000 m / z, ion source temperature of 120 °C, desolvation temperature of 500 °C, capillary voltages of 2.5 kV and -2.2 kV in positive and negative ion modes, respectively, sampling cone voltage of 30 V, collision energy of 10 V to 50 V, nitrogen as the cone gas, and a cone gas flow rate of 50 L / h. The collision gas was argon, the desolvation gas flow rate was 800 L / h, and the calibrator was leucine enkephalin (200 ng / μL), with [M+H]... + (Mass-to-charge ratio of 556.2771) or [MH] - (Mass-to-charge ratio of 554.2615) was used for calibration. Data from the dicarboxylic acid standard series and metabolomics data were obtained after analysis using HMDB, METLIN, Progenesis QI, and MasslynxT.

[0090] 4. Data Analysis Based on data from a series of dicarboxylic acid standards, metabolomics data were subjected to spectral matching and peak extraction to determine endogenous dicarboxylic acids and their retention times under actual testing conditions (e.g., ...). Figure 4 (See the left figure in b). To determine whether the identified endogenous dicarboxylic acids correspond one-to-one with the dicarboxylic acid standard series in terms of the number of carbon atoms, as shown in Table 3, malonic acid, eicosanoic acid, and hexacosanoic acid were not detected due to weak detection peaks.

[0091] Figure 5This diagram shows the extraction chromatograms of various biological matrix samples and the elution sequence of endogenous dicarboxylic acids and analyte metabolites. Figure 5 In figures a to d, the black curves in each figure schematically show the extraction chromatograms of human serum and human plasma, rabbit liver tissue or A549 cells, and the red curves schematically show the extraction chromatogram peaks of endogenous dicarboxylic acid series compounds. Figure 5 e schematically illustrates the analyte i, which is located in the chromatographic elution order between endogenous succinic acid (m=4 carbon atoms) and endogenous glutaric acid (n=5 carbon atoms).

[0092] The regression equation shown in Equation (1) is constructed based on the standard retention times of a series of dicarboxylic acid standards under standard test conditions. The coefficients of the regression equation are fitted according to the detected endogenous dicarboxylic acids and their retention times under actual test conditions, resulting in the prediction equation shown in Equation (2). Based on the prediction equation shown in Equation (2), the retention times of malonic acid, eicosanoic acid, and hexadecanoic acid under actual test conditions are predicted, and all 25 endogenous dicarboxylic acids and their retention times under actual test conditions are output.

[0093] Based on the retention time of endogenous dicarboxylic acid and the analyte metabolite i under actual test conditions and the retention index shown in formula (3), the retention index of the analyte metabolite i under actual test conditions is calculated. Based on the retention index of the analyte metabolite i under actual test conditions and the standard retention time of the dicarboxylic acid standard under standard test conditions, the retention time of the analyte metabolite i under actual test conditions is corrected to the retention time under standard test conditions.

[0094] Figure 6 This describes the changes in retention time deviations of various analyte metabolites before and after correction under different actual testing conditions. See also Figure 6 , Figure 6 Figures A through K show the results for 121 representative metabolites, such as serine, arginine, glutathione, and palmitic acid, in the ranges M1 through M2. 11 These 11 retention time deviations under actual test conditions are the differences between the measured retention times of each analyte metabolite in the biological matrix under different actual test conditions and their standard retention times (which can be pre-determined through ultra-high performance liquid chromatography-mass spectrometry analysis under standard test conditions or obtained by querying a standard database). Black circles represent the retention time deviation before correction, and red circles represent the retention time deviation after correction. Figure 6 After retention time correction, the retention time deviation of most of the metabolites to be tested was significantly reduced from A to K.

[0095] After correction, the average retention time deviations for these 121 analyte metabolites were 1.3 seconds, 1.0 seconds, 1.8 seconds, 1.4 seconds, 0.7 seconds, 1.7 seconds, 1.0 seconds, 0.8 seconds, 0.8 seconds, 5.0 seconds, and 4.7 seconds, respectively. The maximum retention time deviations among them were 6.6 seconds, 5.4 seconds, 9.6 seconds, 6.0 seconds, 5.4 seconds, 8.4 seconds, 5.4 seconds, 5.4 seconds, 10.8 seconds, 25.8 seconds, and 22.8 seconds, respectively. ±0.2 minutes (i.e., ±12 seconds) is generally considered the "golden limit" for retention time drift. Figure 6 It is known that the retention time deviation of 90% of the analytes is within ±0.2 minutes (i.e. ±12 seconds), which means that the retention time correction method of the present invention not only meets the qualitative verification requirements of the world's mainstream and stringent regulations, but also, compared with the traditional retention index correction system, it can effectively correct the retention time data of the metabolites collected in negative ion mode without adding exogenous compounds, and the corresponding retention time coverage ranges from 0.5 minutes to 11.0 minutes.

[0096] Figure 7 This describes the effectiveness of the retention time correction method in correcting retention time deviations of analyte metabolites obtained under actual M3 testing conditions. See also... Figure 7 , Figure 7 Samples a, b, and c were taken from rabbit liver tissue, while samples d, e, and f were taken from human serum. Figure 7 Figures a and d show the comparison between the uncorrected M3 dataset and the uncorrected M0 dataset, where the M3 and M0 datasets are the uncorrected retention time datasets obtained under actual test condition M3 and standard test condition M0, respectively. Figures b and e show the comparison between the corrected M3 dataset and the corrected M0 dataset, where the M3 and M0 datasets are the corrected retention time datasets obtained under actual test condition M3 and standard test condition M0, respectively. Figures c and f show the number of analytes before and after correction at different retention time deviations.

[0097] Depend on Figure 7 It can be seen that when switching from the standard test condition M0 to the actual test condition M3, the retention time deviation of metabolites in rabbit liver tissue and human serum both reached 0.9 min (e.g., Figure 7 The x-coordinates of c and f are shown below.

[0098] Before calibration, such as Figure 7 As shown in a and d, the linear relationship between the M3 uncorrected dataset and the M0 uncorrected dataset is limited (coefficient of determination R² < 0.99). Less than 50% of the analyte metabolites obtained retention times under the actual test condition M3 that match their retention times under the standard test condition M0 (i.e., retention time deviation). ΔtR <0.2 min), this from Figure 7 The data in 'a' and 'd' that deviate from the middle diagonal line can be seen from this.

[0099] After correction, such as Figure 7 As shown in Figures b and e, a good linear relationship is observed between the M3-corrected dataset and the M0-corrected dataset (the coefficients of determination are R0 and R0, respectively). 2 =0.9997 and R 2 =0.9998), the M3 and M0 correction datasets almost entirely cover the middle sloping line, indicating good consistency between them. The retention time deviation of all analyte metabolites is also considered. Δt R Less than 0.2 minutes, such as Figure 6 All the red line segments shown in C and F are distributed along the dashed lines (retaining time deviation). Δt R =0.2min) on the left side.

[0100] Figure 8 This describes the effectiveness of the retention time correction method in correcting retention time deviations of analyte metabolites across different batches and platforms. See also: Figure 8 a, AU, and AC represent the analyte metabolites from human serum before and after correction, respectively, while BU and BC represent the retention time deviations of different analyte metabolites from rabbit liver tissue before and after correction, respectively. Figure 8 b and c show the retention time deviations of different metabolites in 100 repeated injections of human plasma samples before and after correction. Figure 8 Data acquisition for all samples a through c was performed using the first ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) system. Figure 8 In the figure, SU and SC represent the retention time deviations of the analyte metabolites from human serum samples collected by the third UHPLC-MS system before and after correction, respectively, while TU and TC represent the retention time deviations of the analyte metabolites from human serum samples collected by the fourth UHPLC-MS system before and after correction.

[0101] Depend on Figure 8 As can be seen from a, for rabbit liver tissue samples and human serum samples, the retention time deviation of most metabolites between different batches decreased from 3s to 2s. Figure 8 From points b and c, it can be seen that for human plasma samples from 100 repeated injections, the retention time deviation between the same batch is reduced from 12.5s to 8s, applicable to 99% of the analyte metabolites. Figure 8As can be seen from d, for data from three different UHPLC-MS coupling systems, the retention time deviation between the different UHPLC-MS coupling systems was reduced from 1.0 min to 0.2 min, which is applicable to 95% of the metabolites to be tested.

[0102] Therefore, the retention time correction method provided by this invention makes it possible to comprehensively analyze data from different UHPLC-MS systems and different batches. The corrected retention times, provided they meet the standard chromatographic conditions in the standard database, can also be used to reliably identify metabolites. Furthermore, through correction, databases from different UHPLC-MS systems can be interconverted, allowing for combined use for metabolite identification.

[0103] It is worth noting that this retention time correction method is more suitable for reversed-phase columns than hydrophilic columns. Furthermore, since only a few analytes elute at the very beginning of the chromatogram (before the carbon number reaches 0), such as lysine with a retention index set to -100, etc. Figure 8 As shown in Figure a. The retention index of lysine is a manually set value to facilitate correction of its retention time. In summary, the retention time deviation of the analyte metabolites within the same batch is less than 0.1 min (e.g., ...). Figure 8 As shown in Figure a), when collecting data using both positive and negative ion modes, the retention time correction method under negative ion mode can be used to correct the retention time of metabolites.

[0104] Finally, it should be noted that the above embodiments are merely specific implementations of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A retention time correction method for ultra-high performance liquid chromatography-mass spectrometry analysis in negative ion mode, characterized in that, include: S10: Based on the data of a series of dicarboxylic acid standards with sequentially increasing carbon atom numbers and the metabolomics data of the target biological matrix, determine the endogenous dicarboxylic acid series corresponding to the dicarboxylic acid standards in the target biological matrix, and determine the retention time of each endogenous dicarboxylic acid under actual test conditions. The data of each dicarboxylic acid standard is the data obtained by ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) under standard test conditions, including its standard retention time, precise mass-to-charge ratio, and fragment ion structure. The metabolomics data is the data obtained by ultra-high performance liquid chromatography-mass spectrometry (UHPLC-MS) under negative ion mode of the target biological matrix under the actual test conditions, including the retention time, precise mass-to-charge ratio, and fragment ion structure of the endogenous dicarboxylic acid and each metabolite to be tested. S20: Using the retention times of two endogenous dicarboxylic acids with adjacent carbon atom numbers as a reference, calculate the retention index of each analyte metabolite whose retention time falls between the two; and S30: Based on the retention index of each of the metabolites to be tested, correct the retention time of each of the metabolites to be tested under the actual test conditions to the retention time under the standard test conditions.

2. The retention time correction method as described in claim 1, characterized in that, The series of dicarboxylic acid standards are a series of saturated straight-chain alkyl dicarboxylic acids with 2 to 26 carbon atoms, increasing sequentially.

3. The retention time correction method as described in claim 1, characterized in that, Step S10 specifically includes: S11: Based on the data of the aforementioned dicarboxylic acid standard series, perform spectral matching and peak extraction on the metabolomics data to determine the retention times of each endogenous dicarboxylic acid under the actual test conditions. S12: Determine whether the identified endogenous dicarboxylic acid corresponds one-to-one with the dicarboxylic acid standard series in terms of the number of carbon atoms. If so, proceed to S13: Output all identified endogenous dicarboxylic acids and their retention times under the actual test conditions. If not, proceed to S14: Using the number of carbon atoms in the dicarboxylic acid standard as the independent variable and the standard retention time of the dicarboxylic acid standard under the standard test conditions as the dependent variable, construct a cubic regression equation. S15: Using the determined endogenous dicarboxylic acid and its retention time under the actual test conditions as data points, the regression equation is fitted with coefficients to obtain a predictive equation. The predictive equation is a cubic equation, with the number of carbon atoms in the endogenous dicarboxylic acid as the independent variable and the retention time of the endogenous dicarboxylic acid under the actual test conditions as the dependent variable. S16: Based on the prediction equation, predict the retention time of the undetected endogenous dicarboxylic acids in the endogenous dicarboxylic acid series under the actual test conditions, and output all the endogenous dicarboxylic acids and their retention times under the actual test conditions.

4. The retention time correction method as described in claim 3, characterized in that, The regression equation is expressed by the polynomial shown in equation (1): t R sd = 0.0008 L c 3 – 0.0457 L c 2 +1.1828 L c – Equation (1) 2.4883 in, t R sd The standard retention time of the dicarboxylic acid standard under the standard test conditions; L c The number of carbon atoms in the dicarboxylic acid standard.

5. The retention time correction method as described in claim 4, characterized in that, The prediction equation is expressed by the polynomial shown in equation (2): t R = 0.0007 L c 3 – 0.0426 L c 2 +1.1391 L c – Formula 2.2965 (2), in, t R The retention time of the endogenous dicarboxylic acid under the actual test conditions; L c The number of carbon atoms in the endogenous dicarboxylic acid is denoted as .

6. The retention time correction method as described in claim 1, characterized in that, In step S20, the retention index of each of the metabolites to be tested is calculated using the formula shown in equation (3): Equation (3), in, Under the actual test conditions, these refer to the retention time of the analyte metabolite i and the retention times of the two endogenous dicarboxylic acids whose carbon atoms are m and n respectively adjacent to the analyte metabolite i in the elution sequence. ; These are, respectively, the retention index of the metabolite i to be tested, and the retention indices of the two endogenous dicarboxylic acids that are adjacent to the metabolite i in the efflux sequence, with the numbers of carbon atoms being m and n, respectively, under the actual test conditions.

7. The retention time correction method as described in claim 6, characterized in that, The retention index of the endogenous dicarboxylic acid is set to 100 times the number of carbon atoms in the endogenous dicarboxylic acid.

8. The retention time correction method as described in claim 6, characterized in that, In step S30, the retention time of each of the analyte metabolites under the standard test conditions is corrected using the formula shown in equation (4): Equation (4) in, The corrected retention time of the analyte i under the standard test conditions. The standard retention times are those of two dicarboxylic acid standards, m and n, respectively, whose carbon atoms are adjacent to the elution sequence of the metabolite i. .

9. The retention time correction method as described in claim 1, characterized in that, The chromatographic parameters for both the standard test conditions and the actual test conditions include the elution gradient, mobile phase flow rate, column type, and column temperature.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the retention time correction method as described in any one of claims 1 to 9.

Images