Method for realizing automatic extraction of compound information in magnetic resonance mass spectrum

Abstract

The invention provides a method for realizing automatic extraction of compound information in magnetic resonance mass spectrometry, which belongs to the field of quality control of natural product extracts and comprises the steps of original data information conversion, ion information self-adaptive integration, compound chromatogram construction and compound chromatogram outflow characteristic information extraction. Firstly, originally collected signals are converted into data in a NetCDF format, information of all components in each mass spectrum is reserved, and then high-precision mass spectra, retention time and ion abundance information of all ions in a sample are obtained; then, according to the information of the ion fengdu and the high-precision mass spectrum, through a DBSCAN clustering algorithm, obtaining potential valuable characteristic ion center points, and clustering all ions; constructing a chromatogram by the ions in each class according to a retention time sequence; a dynamic window is combined with a continuous smoothing strategy to extract a compound chromatographic peak in each chromatogram, information such as compound retention time, high-precision m / z and peak area is obtained, and finally extraction of compound information in a sample is achieved.

Description

Technical Field

[0001] The invention relates to a method for extracting material information of natural monomer fragrance raw materials in liquid chromatography coupled to magnetic resonance mass spectrometry (LC-MRMS), belongs to the field of quality control of natural product extracts, and particularly relates to a method for realizing automatic extraction of compound information in magnetic resonance mass spectrometry. Background Art

[0002] Natural product extracts are common additives in the production process of cigarettes, which give them rich sensory qualities. Due to the diverse sources of raw materials and complex extraction processes, the chemical composition of natural products is extremely complex. How to achieve quality control of natural products and ensure the stability of the quality of end products is not only a problem faced by the tobacco industry, but also a problem faced by many other industries such as the field of traditional Chinese medicine. The key lies in how to accurately extract and analyze the material information in natural products. In recent years, the field of omics based on chromatographic mass spectrometry has been widely used in food, environment, traditional Chinese medicine and other fields because it can integrate the advantages of separation and identification in the dimensions of chromatography and mass spectrometry and provide a large amount of chemical information. Among them, UPLC-QTOF technology is the most common. However, due to the large number of structural analogs in natural product extracts and the extremely similar molecular weights of compounds, QTOF is challenged. High-resolution equipment based on magnetic resonance mass spectrometry technology provides a new solution for the analysis of complex job samples through ultra-high resolution. However, how to analyze the complex signals generated by the equipment, accurately extract the information of the compounds from them, and realize the analysis of the material composition in natural product extracts is a problem that needs to be solved urgently.

[0003] At present, data analysis methods for high-resolution mass spectrometry such as XCMS, MS-DIAL, MZmine, etc. have been widely used in practical applications. However, in actual analysis, the above methods have problems such as incomplete compound extraction and a large amount of false positive information in chromatographic peak extraction. In addition, the above methods are relatively rarely used in the analysis of material information in natural product extracts. At present, there is no method specifically for the analysis of compound information in LC-MRMS. Therefore, the field urgently needs a technical solution that can solve the above problems.

[0004] The information disclosed in this background technology section is only intended to enhance the understanding of the overall background of the invention and should not be regarded as an acknowledgment or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0005] The purpose of the present invention is to provide a method for analyzing compound information in LC-MRMS.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A method for automatically extracting compound information from magnetic resonance mass spectrometry, comprising:

[0008] Step 1: Raw data information conversion;

[0009] Step 2: Adaptive clustering of ion information;

[0010] Step 3: Compound chromatogram construction;

[0011] Step 4: Extraction and annotation of compound chromatographic elution feature information.

[0012] Optionally, the steps 1 to 4 are specifically as follows:

[0013] First, the original collected signal is converted into data in NetCDF format, and the information of all components in each mass spectrum is retained, so as to obtain high-precision mass spectrum, retention time, and ion abundance information of all ions in the sample; then, based on the ion abundance information and high-precision mass spectrum information, the DBSCAN clustering algorithm is used to obtain the center points of potentially valuable characteristic ions, and all ions are clustered to achieve adaptive classification; the ions in each class are constructed into a chromatogram in the order of retention time; the dynamic window combined with the continuous smoothing strategy is used to extract the compound chromatographic peaks in each chromatogram, and the compound retention time, high-precision m / z, peak area and other information are obtained; according to the isotope of the ion, the adduct ion and water loss, the compound information in the sample is finally extracted.

[0014] Optionally, the method further includes analyzing the liquid chromatography coupled to magnetic resonance high-resolution mass spectrometry data, specifically including the following steps:

[0015] S1: Extraction of chemical components from natural monomer fragrance raw materials for tobacco:

[0016] Weigh an appropriate amount of natural monomer fragrance raw material, place it in a stoppered conical flask, add a suitable extraction solvent, and perform ultrasonic extraction for 10-30 minutes to obtain a natural monomer fragrance raw material extract;

[0017] S2: Sample preparation:

[0018] The extract was shaken and transferred to a centrifuge tube for centrifugation. The supernatant was filtered through a microporous filter membrane, and the filtrate was used as a sample solution for analysis by a high performance liquid chromatography-magnetic resonance mass spectrometer.

[0019] S3: High performance liquid chromatography-magnetic resonance mass spectrometry analysis:

[0020] HPLC-MR-MS analysis: select a suitable chromatographic column and set the HPLC conditions, including column temperature, flow rate, injection volume, mobile phase, etc.; set the magnetic resonance mass spectrometer parameters at the same time, including ion source type, ionization mode, mass axis range, capillary voltage, end plate voltage, nebulizer gas pressure, drying gas flow rate, dryer temperature, number of scans, time domain data set size and resolution; perform sample injection analysis to obtain information about the compounds in the sample.

[0021] S4: For the data in step S3, the RAW format data collected by the instrument is converted into NetCDF general format data, the compound information in the original data is retained, and then high-throughput and accurate extraction of material information is achieved through ion information clustering, chromatogram construction and automatic identification of compound chromatographic peaks.

[0022] Optionally, the specific process of achieving high-throughput and accurate extraction of substance information in step S4 is:

[0023] S41: Data format conversion: Use the operating software on the magnetic resonance mass spectrometer to convert the RAW format data collected from the original data into NetCDF general format data. During the conversion process, set each spectrum to retain 500 ion information.

[0024] S42: Adaptive clustering of ion information: Place the ions under each spectrum into an ion data set A, which stores all ion information in matrix form, where each row contains the information of an ion in the format of [scan, RT, m / z, intensity]. Scan represents the scanning point, RT is the retention time corresponding to the scanning point, m / z is the high-precision charge-to-mass ratio of the ion, and intensity is the intensity of the ion. Based on m / z and intensity, clustering is performed using the DBSCAN algorithm, and the m / z threshold is set to 20ppm. The DBSCAN algorithm extracts the center points of all classes in an iterative manner: first, the ion with the largest intensity is selected as the first center point of the class, and the ions that meet the clustering m / z threshold are divided into the class; then, the point with the largest intensity is found from the unclassified ions as the center point of the new class, and the remaining ions are classified; this process is iterated back and forth until all classes are classified. Subsequently, the clustered ions are re-optimized, and the clustering of the ions to the center points of each class is calculated. For ions that can be reached by multiple classes, they are divided according to the minimum distance.

[0025] S43: Compound chromatogram construction: For each adaptive clustering result, initialize a zero vector (N×1, N is the number of sampling points) of the same length as the sampling point. Place the ion scanning time at the corresponding position of the zero vector, and place the ion intensity into the vector to achieve the construction of the chromatogram (EIC) in this class.

[0026] S44: Extraction of compound chromatographic elution characteristic information: Use wavelet transform to construct a set of smoothing vectors w1, w2, ... wI, where the subscript i represents different smoothing scales, increasing from 0-2 in steps of 0.1. For each EIC signal x, use convolution operation to smooth the signal to obtain xw1, xw2, ..., xwI. For each smoothed signal, obtain the maximum value and mark the position. Connect the maximum value positions of the signals between different smoothing scales in series to construct the maximum value ridge line. Keep the maximum values ​​with a ridge length of more than 10, estimate the signal-to-noise ratio of the maximum value position, use the signal-to-noise ratio of 3 as the threshold, eliminate the maximum values ​​with low signal-to-noise ratio, and finally obtain the chromatographic peak with a reliable signal-to-noise ratio. According to the high-precision m / z value of the ion at the peak maximum, obtain the accurate mark of the chromatographic peak (RT, m / z, peak area, peak height). The compound ions are annotated according to the compound isotope ions (+1, +2, +3, etc.), adduct ions (+Na+, +K+, +NH4+), etc., and finally the compound information is extracted.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention provides a method for automatically extracting compound information from magnetic resonance mass spectrometry, converting the original collected signal data into NetCDF format for structured storage. All component information in each mass spectrum is retained to ensure that the high-precision mass spectrum, retention time and ion abundance information of the ions in each sample are retained. Based on the ion abundance and high-precision mass spectrum information, the DBSCAN clustering algorithm is used to identify potential characteristic ion center points. Clustering all ions and adaptively classifying them helps to better understand the relationship between ions. The ions in each class are sorted by retention time, and the corresponding chromatograms are constructed. This method ensures the timing of information and facilitates subsequent analysis. Using dynamic windows and continuous smoothing strategies, the chromatographic peaks of the compounds are extracted from each chromatogram. Key information such as the retention time, high-precision m / z and peak area of ​​the compound is obtained. Isotope and adduct ion analysis:

[0029] Combined with information such as ion isotopes, adduct ions, and water loss, relevant information about the compounds in the sample can be further extracted.

[0030] Beneficial effects:

[0031] Data structuring: The use of NetCDF format improves the organization and readability of data, facilitating subsequent processing and analysis.

[0032] Segment analysis: DBSCAN clustering can efficiently identify characteristic ions in samples and optimize the data mining process.

[0033] Improve analysis accuracy: Retaining rich information (such as retention time and ion abundance) makes the analysis results more accurate and reliable.

[0034] Visualization and interpretation: The constructed chromatogram helps to intuitively understand the changing trend of sample components, which is convenient for research and communication.

[0035] Deep information extraction: The final extracted compound information provides strong data support for subsequent analysis, research or applications (such as drug development, environmental monitoring, etc.).

[0036] The combination of these steps and effects can effectively support the detailed analysis and understanding of compounds and their characteristics in complex samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0038] Figure 1 A schematic diagram of typical LC-MRMS signals involved in an embodiment of the present invention; wherein a) the LC-MRMS total ion current signal of a natural licorice product extract; b) the mass spectrometry signal at 23.95 min; c) the mass spectrometry data at the scanning point.

[0039] Figure 2 A schematic diagram of constructing a chromatogram using the adaptive clustering and ion collection method of the present invention provided in an embodiment of the present invention; a) original EIC chromatographic signal corresponding to m / z 675.21; b) EIC under different wavelet smoothing scales; c) extracting maximum values ​​under different smoothing scales and constructing wavelet ridges; d) obtaining maximum values ​​corresponding to ridge lengths greater than 10; e) obtaining reliable maximum values ​​under EIC after filtering low signal-to-noise ratios; f) the final chromatographic peak information extraction result.

[0040] Figure 3 An example schematic diagram of implementing EIC peak extraction provided for an embodiment of the present invention; wherein a) the original EIC chromatographic signal corresponding to m / z 675.21; b) EIC under different wavelet smoothing scales; c) extracting the maximum values ​​under different smoothing scales and constructing wavelet ridges; d) obtaining the maximum values ​​corresponding to ridge lengths greater than 10; e) obtaining reliable maximum values ​​under EIC after filtering low signal-to-noise ratios; f) the final chromatographic peak information extraction result.

[0041] Figure 4A schematic diagram of compound ion annotation based on chromatographic peak extraction provided for an embodiment of the invention; wherein a) two chromatographic peaks under EIC are extracted; and b) after annotation, molecular ion peaks and isotope peaks are obtained. DETAILED DESCRIPTION

[0042] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0043] The purpose of the present invention is to provide a method for analyzing compound information in LC-MRMS.

[0044] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0045] Embodiment 1:

[0046] The embodiment of the present invention provides a method for automatically extracting compound information in magnetic resonance mass spectrometry, comprising the following steps:

[0047] S1: Sample preparation process: Weigh 0.5 g (accurate to 0.1 mg) of sample into a 50 mL ground-mouth conical flask, accurately add 10 mL of 50% methanol aqueous solution to dilute and shake well, take an appropriate amount of the dilution, centrifuge at 5000 r / min for 10 min, and take the supernatant for liquid chromatography-magnetic resonance mass spectrometry analysis.

[0048] S2: LC-MRMS analysis:

[0049] The chromatographic conditions for UPLC-QTOF analysis were as follows: mobile phase A: 0.1% formic acid in water, mobile phase B: 0.1% formic acid in acetonitrile. Flow rate: 1.0 mL / min, chromatographic column: Atlantis T3 chromatographic column (4.6×150 mm, 2.7 um). During the chromatographic analysis, the mobile phase gradient was that the mobile phase A accounted for 95% of the total volume of the mobile phase, and the mobile phase B accounted for 5% of the total volume of the mobile phase, which was maintained for 5 minutes, and then decreased to 45% at 30 minutes, decreased to 10% at 55 minutes, and decreased to 0% at 70 minutes and maintained for 10 minutes.

[0050] The mass spectrometry conditions for UPLC-QTOF analysis were as follows: ion source: ESI source; mass spectrometer: FTMS; scanning range: 100-1600Da; scanning mode: positive and negative ion mode; number of scans: 1 time; SIZE: 1M; Accum(s): 0.1s; injection needle flow rate: 2.0uL / min; capillary voltage: 4500V; dryer flow rate: 8.0mL / min; drying temperature: 200℃; nozzle pressure: 1.0bar; flight time: 1.0s.

[0051] S3. Raw data information conversion:

[0052] Under the LC-MRMS platform, the original RAW file type collected by the instrument is directly converted into the universal format data of AIA format to obtain the cdf file corresponding to each sample. During the data conversion process, 500 ion information is selected for each spectrum to be retained.

[0053] S4. Adaptive integration of ion information:

[0054] The ions under each spectrum are placed in an ion data set A, which stores all ion information in matrix form, where each row contains the information of an ion, and its format is [scan, RT, m / z, intensity]. Scan represents the scanning point, RT is the retention time corresponding to the scanning point, m / z is the high-precision charge-to-mass ratio of the ion, and intensity is the ion intensity. Based on m / z and intensity, the DBSCAN algorithm is used for clustering, and the m / z threshold is set to 20ppm. The DBSCAN algorithm extracts the center points of all classes in an iterative manner: first, the ion with the largest intensity is selected as the first center point of the class, and the ions that meet the clustering m / z threshold are divided into this class; then, the point with the largest intensity is found from the unclassified ions as the center point of the new class, and the remaining ions are classified; this process is iterated back and forth until all classes are classified. Subsequently, the clustered ions are re-optimized, and the clustering of the ions to the center points of each class is calculated. For ions that can be reached by multiple classes, they are divided according to the minimum distance.

[0055] S5. Compound chromatogram construction:

[0056] For each adaptive clustering result, a zero vector (N×1, N is the number of sampling points) with the same length as the sampling point is initialized. The ions are placed at the corresponding position of the zero vector according to their scanning time, and the ion intensity is placed in the vector to realize the construction of the chromatogram (EIC) in this class.

[0057] S6. Extraction of compound chromatographic elution characteristic information:

[0058] A set of smoothing vectors w1, w2, ... wI are constructed by wavelet transform, where the subscript i represents different smoothing scales, increasing from 0 to 2 with a step size of 0.1. For each EIC signal x, the signal is smoothed by convolution operation to obtain xw1, xw2, ..., xwI. For each smoothed signal, the maximum value is obtained and the position is marked. The maximum value positions of the signals between different smoothing scales are connected in series to construct the maximum value ridge line. The maximum values ​​with a ridge length of more than 10 are retained, and the signal-to-noise ratio of the maximum value position is estimated. The maximum value with a signal-to-noise ratio of 3 is used as the threshold to eliminate the maximum value with a low signal-to-noise ratio, and finally the chromatographic peak with a reliable signal-to-noise ratio is obtained. According to the high-precision m / z value of the ion at the peak maximum, the accurate labeling of the chromatographic peak is obtained (RT, m / z, peak area, peak height). According to the compound isotope ions (+1, +2, +3, etc.), the summed ions (+Na+, +K+, +NH4+), etc., the compound ions are annotated to finally realize the extraction of compound information.

[0059] Some of the analytical data results of the present invention are as follows:

[0060] The working principle of the present invention is further described. Figure 1 The typical LC-MRMS signals of the present invention include: Illustration a) is the LC-MRMS total ion current signal of the natural licorice product extract; Illustration b) is the mass spectrum signal collected at 23.95min; Illustration c) is the mass spectrum data at a certain scanning point.

[0061] As attached Figure 2 As shown, the chromatogram is constructed by using the adaptive clustering collection of ions in the method of the present invention. a) The original collection of ion distribution; b) The ion clustering result with the center point of m / z 823.414216 based on DBSCAN; c) EIC construction based on the ion clustering effect.

[0062] As attached Figure 3 As shown in the figure, based on the construction of the chromatogram, the present invention realizes an example of EIC peak extraction. a) The original EIC chromatographic signal corresponding to m / z 675.21; b) EIC under different wavelet smoothing scales; c) Extracting the maximum values ​​under different smoothing scales and constructing the wavelet ridge line; d) Obtaining the maximum value corresponding to the ridge line length greater than 10; e) Obtaining the reliable maximum value under EIC after filtering the low signal-to-noise ratio; f) The final chromatographic peak information extraction result.

[0063] As attached Figure 4 As shown in the figure, a) two chromatographic peaks under EIC are extracted; b) after annotation, the molecular ion peak and isotope peak are obtained, realizing the accurate identification of isotope ions.

[0064] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0065] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A method for automatically extracting compound information from magnetic resonance mass spectrometry, characterized in that: include: Step 1: Raw data information conversion; Step 2: Adaptive clustering of ion information; Step 3: Compound chromatogram construction; Step 4: Extraction and annotation of compound chromatographic elution feature information.

2. The method for automatically extracting compound information from magnetic resonance mass spectrometry according to claim 1, characterized in that: The steps one to four are specifically as follows: First, the original collected signal is converted into data in NetCDF format, and the information of all components in each mass spectrum is retained, so as to obtain high-precision mass spectrum, retention time, and ion abundance information of all ions in the sample; then, based on the ion abundance information and high-precision mass spectrum information, the DBSCAN clustering algorithm is used to obtain the center points of potentially valuable characteristic ions, and all ions are clustered to achieve adaptive classification; the ions in each class are constructed into a chromatogram in the order of retention time; the dynamic window combined with the continuous smoothing strategy is used to extract the compound chromatographic peaks in each chromatogram, and the compound retention time, high-precision m / z, peak area and other information are obtained; according to the isotope of the ion, the adduct ion and water loss, the compound information in the sample is finally extracted.

3. The method for automatically extracting compound information from magnetic resonance mass spectrometry according to claim 1, characterized in that: It also includes analysis of liquid chromatography coupled to magnetic resonance high-resolution mass spectrometry data, specifically The following steps are involved: S1: Extraction of chemical components from natural monomer fragrance raw materials for tobacco: Weigh an appropriate amount of natural monomer fragrance raw material, place it in a stoppered conical flask, add a suitable extraction solvent, and perform ultrasonic extraction for 10-30 minutes to obtain a natural monomer fragrance raw material extract; S2: Sample preparation: The extract was shaken and transferred to a centrifuge tube for centrifugation. The supernatant was filtered through a microporous filter membrane, and the filtrate was used as a sample solution for analysis by a high performance liquid chromatography-magnetic resonance mass spectrometer. S3: High performance liquid chromatography-magnetic resonance mass spectrometry analysis: HPLC-MR-MS analysis: Select a suitable chromatographic column and set the HPLC conditions, including column temperature, flow rate, injection volume, mobile phase, etc.; set the MR-MS parameters, including ion source type, ionization mode, mass axis range, capillary voltage, end plate voltage, nebulizer gas pressure, drying gas flow rate, dryer temperature, number of scans, time domain data set size and resolution; Inject sample for analysis to obtain information about compounds in the sample; S4: For the data in step S3, the RAW format data collected by the instrument is converted into NetCDF general format data, the compound information in the original data is retained, and then high-throughput and accurate extraction of material information is achieved through ion information clustering, chromatogram construction and automatic identification of compound chromatographic peaks.

4. The method for automatically extracting compound information from magnetic resonance mass spectrometry according to claim 3, characterized in that: The specific process of achieving high-throughput and accurate extraction of substance information in step S4 is as follows: S41: Data format conversion: Use the operating software on the magnetic resonance mass spectrometer to convert the RAW format data collected from the original data into the NetCDF general format data. During the conversion process, set each spectrum to retain 500 ion information; S42: Adaptive clustering of ion information: Place the ions under each spectrum into an ion data set A, which stores all ion information in matrix form, where each row contains the information of an ion in the format of [scan, RT, m / z, intensity]; scan represents the scanning point, RT is the retention time corresponding to the scanning point, m / z is the high-precision charge-to-mass ratio of the ion, and intensity is the intensity of the ion; clustering is performed using the DBSCAN algorithm based on m / z and intensity, and the m / z threshold is set to 20ppm; the DBSCAN algorithm extracts the center points of all classes in an iterative manner: first, the ion with the largest intensity is selected as the first center point of the class, and the ions that meet the clustering m / z threshold are divided into the class; then the point with the largest intensity among the unclassified ions is found as the center point of the new class, and the remaining ions are classified; This process is repeated until all sub-classes are classified; then, the clustered ions are re-optimized, and the clustering of ions to the center of each class is calculated. For ions that can be reached by multiple classes, they are divided according to the minimum distance; S43: Compound chromatogram construction: For each adaptive clustering result, initialize a zero vector (N×1, N is the number of sampling points) of the same length as the sampling point; place the ion at the corresponding position of the zero vector according to the scanning time, and place the ion intensity into the vector to realize the construction of the chromatogram (EIC) in this class; S44: Extraction of compound chromatographic elution characteristic information: Use wavelet transform to construct a set of smoothing vectors w1, w2, ... wI, where the subscript i represents different smoothing scales, increasing from 0 to 2 with a step size of 0.1; for each EIC signal x, use convolution operation to smooth the signal to obtain xw1, xw2, ..., xwI; for each smoothed signal, obtain the maximum value and mark the position, connect the maximum value positions of signals between different smoothing scales in series to construct the maximum value ridge line; retain the maximum values ​​with a ridge line length of more than 10, estimate the signal-to-noise ratio of the maximum value position, use the signal-to-noise ratio of 3 as the threshold, eliminate the maximum values ​​with low signal-to-noise ratio, and finally obtain the chromatographic peak with a reliable signal-to-noise ratio; obtain the accurate marking of the chromatographic peak based on the high-precision m / z value of the ion at the peak maximum (RT, m / z, peak area, peak height); annotate compound ions according to compound isotope ions (+1, +2, +3, etc.), adduct ions (+Na+, +K+, +NH4+), etc., and finally realize the extraction of compound information.

Images