Oxidized fatty acid large-scale qualitative method based on non-targeted metabonomics

Abstract

The invention relates to an oxidized fatty acid large-scale qualitative method based on non-targeted metabonomics. The method comprises the following steps: firstly, enriching oxidized fatty acid in a biological sample to obtain non-targeted metabonomics data; secondly, constructing a self-built database which comprises an oxidized fatty acid mass spectrum library, an oxidized fatty acid structure library and an in silico oxidized fatty acid virtual structure library; and then, carrying out self-established virtual database search on data of an oxidized fatty acid extraction sample, and carrying out large-scale identification on the oxidized fatty acid by comparing a first-level ion mass-to-charge ratio, a second-level mass spectrum and / or second-level ion characteristic fragments, retention time and an in-source cracking mode. And finally, co-constructing a molecular network by the secondary mass spectra recorded in all the databases, the identified secondary mass spectra in the samples and the secondary mass spectra of all the unknown metabolites in the samples based on secondary similarity, and carrying out expansion annotation on the unknown oxidized fatty acid through a metabolic reaction relationship.

Description

Technical Field

[0001] The invention relates to the field of analytical chemistry and is a new method for large-scale qualitative analysis of oxidized fatty acids based on non-targeted metabolomics. Background Art

[0002] Eicosanoids and related oxidized lipids are key small bioactive mediators in organisms. Oxidized fatty acids are a very important class of compounds among oxidized lipids. So far, the types of oxidized fatty acids detected are still very limited, which limits the extensive annotation and in-depth study of this class of compounds in biological systems. Oxidized fatty acids refer to a class of bioactive substances related to immunity or stress that are produced by the oxidation of polyunsaturated fatty acid lipids under conditions such as enzyme action, auto-oxidation or heat treatment. Oxidized fatty acids of plant origin mainly include monohydroxy fatty acids (Hydroxy fatty acids), epoxy fatty acids and hydroperoxyl fatty acid compounds derived from linoleic acid (LA) and linolenic acid (ALA). Oxidized lipids of animal origin mainly include enzymatic oxidation products of arachidonic acid (ARA), eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), including prostaglandins (PGs), leukotrienes (LTs), lipoxins (LXs), thromboxanes (TXs), hydroxyeicosatrienoic acid (EETs) and hydroxyeicosatetraenoic acid (HETEs).

[0003] The analytical method based on the LC-MS platform has been widely used in the separation and analysis of oxidized fatty acids. The existing large-scale qualitative methods of oxidized fatty acid compounds mainly encounter the following bottlenecks. First, previous qualitative studies based on LC-MS / MS mainly focused on one or several key types of oxidized fatty acids, and it was difficult to achieve in-depth annotation of oxidized fatty acids, such as 2- or 3-hydroxy fatty acids from animals or saturated hydroxyl oxidized fatty acids from plants, which have been widely studied. Second, the current large-scale qualitative methods of oxidized fatty acids mainly perform rough qualitative analysis through the prediction and matching of molecular weight or molecular formula, such as directly searching existing databases such as Lipidmap. Qualitative analysis can be performed by the large-scale qualitative software LOBSTAHS. Similar methods only rely on the primary parent ion mass of the metabolite, and the qualitative results may contain many false positive candidates. In addition, in order to improve the detection sensitivity, derivatization reagents are also playing an increasingly important role in the detection of oxidized fatty acids. However, due to the participation of derivatization reagents, the fragment ions in the secondary mass spectrometry are mainly produced by the derivatization reagents. In addition, the specificity of the derivatization reagents in the qualitative process based on the derivatization method is often insufficient, and the results may contain a large number of false positives. In addition, the available oxidized fatty acid standards are expensive and limited in number, and the number of reference spectra of oxidized fatty acids in the database is also very limited. Although some studies have discussed the characteristic fragments of the secondary mass spectrometry of oxidized fatty acids, due to the different content and interference levels in the samples, the characteristic fragments cannot be successfully obtained in many cases. In addition, the method of predicting the retention time of oxidized fatty acids through accurate structural prediction and simulation is often based on the accurate consideration of the functional groups and functional group positions of oxidized fatty acids in order to accurately predict the retention time of oxidized fatty acids. This is not conducive to the annotation of many oxidized fatty acids that can only be predicted at the molecular formula level. From the perspective of secondary mass spectrometry, the oxidized fatty acids in biological samples are extremely low in content and there are a large number of endogenous substances, which will cause serious interference to the analysis of both primary and secondary mass spectrometry, resulting in poor secondary mass spectrometry quality and affecting its successful identification. Finally, due to the widespread isomerism of oxidized fatty acid components in biological samples and the high secondary similarity, it is difficult to distinguish and correctly identify oxidized lipid isomers.For the existing methods of qualitative analysis of oxidized fatty acids using molecular networks, they are mainly large-scale qualitative methods for oxidized fatty acids of animal origin. This method is limited to oxidized fatty acids with very rich secondary fragments, such as arachidonic acid, and is limited to the association of similar molecules based on spectral similarity without in-depth annotation of the network. Although the unknown oxidized fatty acids in the sample are captured by the standard secondary mass spectrometry, the standard spectra of oxidized fatty acids in the existing secondary mass spectrometry library and the extended annotation capabilities of the network are not utilized. The use of the reference secondary mass spectrometry is limited to traditional library search, resulting in low utilization. The secondary similarity of oxidized fatty acids in the sample and the correlation between metabolites in the organism are rarely used for structural inference and annotation of oxidized fatty acids. The network itself has no structural analysis capability, resulting in its structural annotation being limited to standard sample data matching, and the annotation effect is poor. Therefore, it is urgent to develop a large-scale qualitative method for oxidized lipids, which can enhance the utilization and annotation of its secondary mass spectrometry in terms of qualitative analysis, improve the coverage of qualitative analysis and the reliability of qualitative results, and can be used for comprehensive and in-depth annotation of oxidized fatty acids from multiple sources. Summary of the invention

[0004] The present invention relates to a novel method for large-scale qualitative analysis of oxidized fatty acids based on non-targeted metabolomics. First, oxidized fatty acids in biological samples are enriched, and non-targeted metabolomics data in negative ion mode are collected using ultra-high performance liquid chromatography-high resolution mass spectrometry; secondly, an oxidized fatty acid ultra-high performance liquid chromatography-high resolution mass spectrometry self-built database is constructed, including: an oxidized fatty acid mass spectral library (including molecular formula, accurate mass number and [MH]-, predicted retention time, experimental MS 2 The oxidized fatty acids were identified by comparing the mass-to-charge ratio of the primary ion, the secondary mass spectra and / or the characteristic fragments and retention time as well as the in-source fragmentation pattern. Finally, the molecular network was constructed based on the secondary mass spectra of all the databases, the identified secondary mass spectra in the samples, and the secondary mass spectra of all the unknown metabolites in the samples, and the unknown oxidized fatty acids were annotated with the metabolic reaction relationships. Finally, the large-scale qualitative analysis of oxidized fatty acids is completed. The technical solution steps adopted by the present invention are as follows:

[0005] The first step is to pre-treat the biological sample to be tested and extract the oxidized fatty acid components in the sample to be tested;

[0006] In the second step, ultra-high performance liquid chromatography-high resolution mass spectrometry was used to perform non-targeted metabolome analysis on the pre-treated oxidized fatty acid extracts, and the liquid chromatography retention time, primary mass spectrometry information including the mass-to-charge ratio and intensity information of the primary ions, and secondary mass spectrometry information including the mass-to-charge ratio and intensity information of the secondary ions were collected; the primary ions here refer to the ions collected directly after the compounds are ionized; the secondary ions refer to the ions collected after the compounds are ionized and then collided and fragmented with a certain energy;

[0007] The third step is to purchase actual oxidized fatty acid standards covering chemical structures with carbon atoms between 10 and 30, oxygen atoms between 2 and 15, and double bonds between 2 and 15 and not less than 50 (e.g., 12-hydroxydodecanoic acid, 3-hydroxydecanoic acid, Prostaglandin D2, (+ / -)12(13)-EpOME, 9-HODE, etc., 54) and obtain their corresponding chromatographic retention times to construct a structural retention time relationship and regression equation for the carbon number, total number of double bonds, and oxygen number in the oxidized fatty acid molecular formula relative to the retention time: predicted retention time = k1* carbon number of fatty acid + k2* unsaturation + k3* oxygen number, where k is the obtained model parameter;

[0008] In the fourth step, the difference between the primary parent ion and the secondary fragment ion is called neutral loss. The secondary fragmentation pattern of oxidized fatty acids in negative ion mode is: 1) neutral loss of carbon dioxide (CO2), 2) neutral loss of water (H2O) and carbon dioxide (CO2), 3) m / z 59.0138 characteristic fragment of acetic acid and neutral loss of 46.0049Da, 4) neutral loss of acetic acid at 46.0049Da and neutral loss of water, which meets one of the four conditions;

[0009] The fifth step is to construct a self-built database of oxidized fatty acid ultra-high performance liquid chromatography-high resolution mass spectrometry. The database includes three sub-databases: oxidized fatty acid mass spectrum library, oxidized fatty acid structure library, and in silico oxidized fatty acid virtual structure library. The specific construction method is as follows;

[0010] 1) Construction of oxidized fatty acid mass spectral library: Collect oxidized fatty acid information from mass spectral databases (such as the Global Natural Products Social Molecular Networking (GNPS) mass spectral database) and literature reports (such as Cell Chem Biol. 2019, 26(3): 433-442) to construct an oxidized fatty acid mass spectral library. The database includes the following information: molecular information of oxidized fatty acids included in the mass spectral library or / and literature, including the name of the oxidized fatty acid, molecular formula, accurate mass number, primary parent ion [MH]-, secondary mass spectral information, and retention time predicted by the model in S3;

[0011] 2) Construction of the oxidized fatty acid structure library: Collect all oxidized fatty acids in the lipid structure database (Lipid Metabolites And Pathways Strategy (LIPDMAPS)) whose molecular formulas contain 10-30 carbon atoms, 2-15 oxygen atoms, and 2-15 double bonds for all lipids of the "Fatty Acyls" category to construct an oxidized fatty acid structure library, which includes the oxidized fatty acid name, molecular formula, accurate mass number, primary ion [MH]-, secondary mass spectrum information, and retention time predicted according to the model in S3;

[0012] 3) Construction of in silico oxidized fatty acid virtual structure library: For molecular formula forms not present in 1) and 2), the molecular formulas were supplemented and expanded within the range of 10 to 30 oxidized fatty acid carbon numbers, 1 to carbon chain length / 2 total number of double bonds, and 3 to carbon chain length / 2 oxygen number, and their corresponding molecular formulas, molecular weights, accurate mass numbers, and corresponding mass-to-charge ratios [MH]-information in negative ion mode, secondary mass spectrometry characteristic fragments predicted in S4, and retention times predicted by the model in S3 were obtained;

[0013] Step 6: Match the LC retention time, primary mass spectrometry information and secondary mass spectrometry information of the metabolites in the oxidized fatty acid extract obtained in step S2 with the oxidized fatty acid mass spectrum library, oxidized fatty acid structure library and in silico oxidized fatty acid virtual structure library in S5 in turn; and infer whether the candidate compound is a hydroxyl-containing oxidized fatty acid by extracting the in-source fragmentation pattern [MH-H2O];

[0014] In the seventh step, all the secondary mass spectra of oxidized fatty acids collected in the oxidized fatty acid mass spectrum library in S5, the secondary mass spectra of oxidized fatty acids collected in S2 and successfully matched in S4, and the secondary mass spectra of all unknown metabolites collected in S2 and not successfully matched in S4 were co-constructed into a molecular network on the GNPS platform based on the similarity of secondary mass spectra. For the unknown oxidized fatty acid molecules that were successfully associated with the molecular network, extended annotation of oxidized fatty acids and their modified metabolites was performed through the collected 63 common metabolic reactions and modification types.

[0015] The method of the present invention has the following beneficial effects and advantages when annotating oxidized fatty acid compounds on a large scale: first, previous studies have mainly focused on one or several types of oxidized fatty acids, and the key molecules of interest have category specificity, such as oxidized fatty acids from animals or plants, which limits the application scenarios of the method. The present invention can identify oxidized fatty acids in animals and plants on a large scale by using a variety of qualitative methods, and the qualitative methods are accurate and reliable; second, the qualitative method of studying oxidized fatty acids in the present invention is different from the traditional derivatization method or relying solely on MS. 1In addition, although the characteristic fragments of some hydroxyl oxidized lipids have been discussed in previous studies, the fatty acids in the samples may lack characteristic fragments in complex matrices, and due to the high heterogeneity of such substances, many secondary spectra of oxidized fatty acid isomers are often highly similar. We can annotate unknown oxidized fatty acids through mass spectrometry fragmentation patterns and spectrum similarity association. In order to accurately characterize the retention time, the functional groups and functional group positions of oxidized fatty acids are often considered, which is very unfavorable for the annotation of many oxidized fatty acids that can only be predicted at the molecular formula level. The present invention uses a structure and retention time prediction model. According to the relationship between structural features and retention time, the retention time can be directly predicted by molecular formula according to the model, avoiding the cumbersome traditional methods of oxidized fatty acids. The method does not involve complex structure prediction simulation processes and analysis of unnecessary structural information, but directly performs molecular formula expansion and retention time prediction based on key structure-related parameters, including the number of C-numbers, degree of unsaturation and number of oxygen-containing atoms, thereby improving the execution efficiency of the method prediction; secondly, the traditional method tries to avoid the in-source cleavage of compounds in the mass spectrometry, while the present method makes full use of the property that hydroxy-oxidized fatty acids are easily fragmented in the ion source to assist in the identification of hydroxy-oxidized fatty acids; at the same time, when the existing methods use molecular networks to characterize oxidized fatty acids, they only perform rough associations and do not perform in-depth annotations on the networks. At the same time, although the existing methods associate unknown oxidized fatty acids in samples through standard secondary mass spectra, they do not use the successfully characterized and existing secondary mass spectrometry libraries for annotation expansion. At the same time, since the network itself has no structural analysis capability, its structural annotation is limited to standard samples and reference spectrum data, resulting in poor annotation effects. The present invention can realize the large-scale annotation of oxidized fatty acid-specific molecular networks by constructing a self-built database of oxidized fatty acids with diversified information. At the same time, the specific network can make full use of existing standards and reference spectra, and associate the rich oxidized fatty acid data annotated de novo in the experimental secondary level with more metabolites with similar structures, while utilizing the potential biological relationships between metabolites for further qualitative characterization of oxidized fatty acids. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 Schematic diagram of the large-scale qualitative process of oxidized fatty acids;

[0017] Figure 2 Construction of a prediction model for retention time of oxidized fatty acids;

[0018] Figure 3 Characteristic fragmentation patterns of oxidized fatty acids;

[0019] Figure 4 In-source fragmentation patterns of different types of oxidized fatty acids;

[0020] Figure 5 The proportion of secondary mass spectrometry sources in the oxidized fatty acid mass spectral library;

[0021] Figure 6 Oxidized fatty acid structure library: Spatial distribution of the number of C atoms, degree of unsaturation and number of oxygen atoms in oxidized fatty acid molecules;

[0022] Figure 7 The spatial distribution diagram of the number of C atoms, unsaturation and oxygen atoms in the oxidized fatty acid molecules of the in silico oxidized fatty acid virtual structure library;

[0023] Figure 8 Extracted ion chromatograms of the molecular formula C16H30O3 m / z 269.2122 in the actual sample and its predicted in-source fragmentation form m / z 251.2017;

[0024] Fig. 9 Schematic diagram of the local network of characteristic molecular networks of oxidized fatty acids. DETAILED DESCRIPTION

[0025] The following is a detailed description of the implementation of the present invention in conjunction with the attached drawings and tables: This embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation method and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

[0026] Example 1

[0027] This method introduces a large-scale qualitative method for oxidized fatty acids, which mainly includes: first, enriching oxidized fatty acids in biological samples, and using ultra-high performance liquid chromatography-high-resolution mass spectrometry to collect non-targeted metabolomics data in negative ion mode; second, constructing an ultra-high performance liquid chromatography-high-resolution mass spectrometry self-built database of oxidized fatty acids, including: oxidized fatty acid mass spectral library (including molecular formula, accurate mass number and [MH]-, predicted retention time, experimental MS 2The oxidized fatty acids were identified by comparing the mass-to-charge ratio of the primary ion, the secondary mass spectra and / or the secondary ion characteristic fragments and the in-source fragmentation pattern [M-H2O-H]-), the oxidized fatty acids structure library (including the molecular formula, accurate mass and [MH]- of the oxidized fatty acids included in LIPIDMAPS, the predicted retention time, the predicted secondary mass spectrometry characteristic fragments and the in-source fragmentation pattern [M-H2O-H]-), and the insilico oxidized fatty acids virtual structure library (including the molecular formula, accurate mass and [MH]-, the predicted retention time, the predicted secondary mass spectrometry characteristic fragments and the in-source fragmentation pattern [M-H2O-H]-); then, the non-targeted metabolite data of the oxidized fatty acids extracted samples were searched in the self-built virtual database of oxidized fatty acids ultra-high performance liquid chromatography-high resolution mass spectrometry, and the oxidized fatty acids were identified on a large scale by comparing the mass-to-charge ratio of the primary ion, the secondary mass spectra and / or the secondary ion characteristic fragments and the retention time and the in-source fragmentation pattern. Finally, the secondary mass spectra included in all databases, the identified secondary mass spectra in the samples, and the secondary mass spectra of all unknown metabolites in the samples were used to build a molecular network based on secondary similarity, and the unknown oxidized fatty acids were extended and annotated through metabolic reaction relationships. Finally, the large-scale characterization of oxidized fatty acids was completed. Figure 1 shown.

[0028] Enrichment and extraction of oxidized fatty acids and sample preparation: Weigh 10 mg of freeze-dried powder of rice leaves and stems, add 1 mL of 0.5 mM butylated hydroxytoluene (BHT), extract at low temperature at 4 ° C for 1 h, centrifuge at 14000 rpm at 4 ° C for 10 min, and freeze-dry the supernatant. Add 200 μL of methanol solution containing 0.5% formic acid by volume to the freeze-dried material, re-dissolve and vortex, add 300 μL of water and 400 μL of ethyl acetate to it, vortex centrifuge, carefully remove the upper layer of ethyl acetate solution with a pipette, add 400 μL of ethyl acetate to the lower layer and repeat the extraction, combine the ethyl acetate layer and freeze-dry, re-dissolve it with 50 μL of 80% methanol water by volume, vortex centrifuge, put it into the injection bottle and put it into the sample tray to wait for injection.

[0029] Non-targeted chromatography-mass spectrometry information acquisition: In order to obtain secondary mass spectrometry data rich in oxidized fatty acids, this experiment used an analytical instrument combined with an ACQUITY UHPLC ultra-high performance liquid chromatography system (UPLC, Waters, Milford, MA, USA.) and a Q Exactive HF high-resolution mass spectrometer (Thermo Fisher Scientific, Rockford, IL, USA.) to collect non-targeted metabolomics data of rice leaf and stem samples.

[0030] The chromatographic analysis conditions were as follows: the liquid chromatography conditions in the positive and negative ion modes were pure acetonitrile and 0.5% formic acid / water (v:v) for phase A and phase B, respectively. The flow rate was 0.35 mL / min, and the total analysis time was 25.1 min. The elution gradient started with 10% B (v:v), linearly increased to 25% B within 5 min, and then continued to increase linearly to 35% B within 5 min. Then, it continued to increase linearly to 75% B within the next 10 min, and quickly increased to 95% B within 0.1 min and maintained until 25 min, and then decreased to 10% B within 0.1 min. The samples were analyzed using an ACQUITY BEH C18 column (100 mm×2.1 mm, 1.7 μm, Waters, Milford, MA, USA). The column temperature was set to 50°C. The injection chamber temperature was 4°C, and the injection volume was 5 μL.

[0031] The mass spectrometry analysis conditions were as follows: the scanning mode was full scan plus secondary mass spectrometry acquisition mode (full MS / data-dependent (ddMS2)). In the full scan mass spectrometry setting, the resolution was 120,000. The automatic gain control target (AGCtarget) and the maximum injection time (maximum IT) were set to 3×106 ion capacity and 100 ms, respectively. The full scan mass scanning range was m / z 85 to 1250. In the secondary mass spectrometry setting, the automatic gain control target (AGC target) and the maximum injection time (maximum IT) were set to 1×10 5 ion capacity and 50ms. The isolation window was m / z 1.0. The collision energy used a mixed normalized energy of 15%, 30% and 45%. The secondary acquisition was triggered by the top 10 ions with the strongest response in each full scan cycle. The scan range was divided into multi-segment scans (85-280, 270-390, 380-560, 550-1250) and multi-needle iterations (4 iterations per needle). All data were collected in positive ion mode. The spray voltage was 3.5 kV, the temperature of the ion transfer tube was 320°C, and the heater temperature of the auxiliary gas was 350°C. The sheath gas and auxiliary gas flow rates were 45 and 10 (in arbitrary units), respectively. The S-lens was set to 50.0 (in arbitrary units).

[0032] The mass spectrometry analysis conditions were as follows: the scanning mode was full scan plus secondary mass spectrometry acquisition mode (full MS / data-dependent (ddMS2)). In the full scan mass spectrometry setting, the resolution was 120,000. The automatic gain control target (AGCtarget) and the maximum injection time (maximum IT) were set to 3×106 ion capacity and 100 ms, respectively. The full scan mass scanning range was m / z 85 to 1250. In the secondary mass spectrometry setting, the automatic gain control target (AGC target) and the maximum injection time (maximum IT) were set to 1×10 5 ion capacity and 50ms. The isolation window was m / z 1.0. The collision energy used a mixed normalized energy of 15%, 30% and 45%. The secondary acquisition was triggered by the top 10 ions with the strongest response in each full scan cycle. The scan range was divided into multi-segment scans (85-280, 270-390, 380-560, 550-1250) and multi-needle iterations (4 iterations per needle). All data were collected in positive ion mode. The spray voltage was 3.5 kV, the temperature of the ion transfer tube was 320°C, and the heater temperature of the auxiliary gas was 350°C. The sheath gas and auxiliary gas flow rates were 45 and 10 (in arbitrary units), respectively. The S-lens was set to 50.0 (in arbitrary units).

[0033] Scale-up qualitative process of oxidized fatty acids: Figure 1 As shown in the figure, first, the oxidized fatty acids in biological samples were enriched, and ultra-high performance liquid chromatography-high resolution mass spectrometry was used to collect non-targeted metabolomics data in negative ion mode; secondly, an ultra-high performance liquid chromatography-high resolution mass spectrometry self-built database of oxidized fatty acids was constructed, including: an oxidized fatty acid mass spectral library (including molecular formula, accurate mass number and [MH]-, predicted retention time, experimental MS 2The oxidized fatty acids were identified by comparing the mass-to-charge ratio of the primary ion, the secondary mass spectra and / or the characteristic fragments and retention time as well as the in-source fragmentation pattern. Finally, the molecular network was constructed based on the secondary mass spectra of all the databases, the identified secondary mass spectra in the samples, and the secondary mass spectra of all the unknown metabolites in the samples, and the unknown oxidized fatty acids were annotated with the metabolic reaction relationships. Finally, the large-scale identification of oxidized fatty acids was completed.

[0034] Establishing the relationship between the structural characteristics of oxidized fatty acids and retention time through standard substances: In order to accurately characterize the retention time, previous studies often considered the functional groups and functional group positions of oxidized fatty acids, which was complicated and required many factors to be considered. In order to directly predict the retention time of oxidized fatty acids from the molecular formula, this method performed multivariate linear regression through standard oxidized fatty acid standards and found that the retention time range of oxidized lipids can be predicted by the carbon chain length, number of oxygen atoms and degree of unsaturation of the oxidized fatty acids, that is, the corresponding retention time range prediction can be obtained directly from the molecular formula, thereby greatly broadening the applicable scope of the retention time. The retention time prediction models for different data sets are as follows: Figure 2 shown.

[0035] like Figure 2 As shown in A, in the present invention, 53 standard samples were collected, 36 of which were used as training data and 18 were used as verification data (see Appendix 1 for detailed information on the standard samples). Multiple linear regression of the number of carbon atoms, the number of oxygen atoms, and the total number of double bonds (i.e., unsaturation) in the oxidized fatty acids relative to the retention time was performed, and the regression equation was retention time = 333.979 + 57.562*C-number-110.82*O-number-42.630*Unsaturation (wherein C-number is the number of carbon atoms, O-number is the number of oxygen atoms, and Unsaturation is the number of double bonds), with an Adjusted R2 of 0.932, an average absolute error of 0.58 min, a median absolute error of 0.39 min, and an average relative error of 4.36%.

[0036] In order to verify the practicality of this method, we used the retention time information of oxidized fatty acids under different chromatographic conditions reported in the literature, and also performed multiple linear regression of oxidized fatty acid C-chain length, oxygen number, and unsaturation relative to retention time. We used the retention time of oxidized fatty acids in three data sets in the literature to perform multiple linear regression of oxidized fatty acid C-chain length, oxygen number, and unsaturation relative to retention time. Data set 1 contained a total of 210 oxidized fatty acid standards, of which 128 were used as training data and 82 as validation data. Figure 2 As shown in B, the regression equation obtained is t R =1053.434+67.812*C-number-324.527*O-number-44.658*Unsaturation, with an Adjusted R2 of 0.843, an average absolute error of 0.38min, a median absolute error of 0.29min, and an average relative error of 14.4%. Dataset 2 contains 152 oxidized fatty acid standards reported in the literature, 102 of which are used as training sets and 50 as validation sets. Figure 2 C, the regression equation is t R =1053.434+67.812*C-number-324.527*O-number-44.658*Unsaturation, R2 is 0.868, the average absolute error is 1.51min, the median absolute error is 1.27min, and the average relative error is 13.40%. Data set 3 contains 150 oxidized fatty acid standards, such as Figure 2 D, the regression equation obtained is t R =416.058+20.842*C-number-103.976*O-number-11.336*Unsaturation, 95 of which were used as training sets and 55 as validation sets, with an R2 of 0.887, an average absolute error of 0.53min, a median absolute error of 0.40min, and an average relative error of 10.0%. The above results show that even under different chromatographic analysis conditions, the structure-retention model under different chromatographic analysis conditions can be constructed by using the universal multivariate linear regression equation of the C-chain length, oxygen number, and unsaturation of the oxidized fatty acid standards relative to the retention time. It should be noted that this method is only applicable to oxidized fatty acids containing only one carboxyl group, and the structure contains only three elements: C, H, and O. If the structure contains more than one carboxyl functional group, it does not follow the structural retention law.

[0037] Characteristic fragmentation patterns of oxidized fatty acids: By summarizing the mass spectrometry patterns of 53 fatty acid standards, standard oxidized fatty acid samples can be mainly divided into the following categories. The first type of structure is arachidonic acid, and typical structures include animal-derived prostaglandins, thromboxanes and leukotriene compounds. This type of component has extremely rich secondary fragments, and the mass spectrometry fragmentation of the fatty acid chain is relatively complete, such as Figure 3 As shown in the figure, the typical fragmentation in the spectrum is the fragmentation of the carboxyl group, which is manifested as one or more continuous neutral losses of water 18 and rich and diverse fragmentation of fatty acid chains and neutral loss of carbon dioxide 44. The second type of structure is arachidonic acid. Typical structures include arachidonic acid from plant sources (also present in animal samples) and hydroxy fatty acid components frequently reported in the literature, such as Figure 3 As shown, compared with the oxidized fatty acids with more double bonds in the structure, their secondary fragments are relatively rich. The most obvious fragmentation characteristics are the neutral loss of 44 of CO2 of the carboxyl group and the neutral loss of 18 of the hydroxyl group in the structure, as well as the fragmentation of the hydroxyl functional group. For the single-chain oxidized fatty acids with fewer double bonds in the structure, such as Figure 3 As shown, its secondary fragments are less abundant, and the neutral loss of CO2 and water is not as abundant as the aforementioned polyunsaturated fatty acids, and its secondary fragmentation mainly occurs near the functional groups.

[0038] Therefore, for oxidized fatty acids with a small number of oxygen-containing functional groups, the most characteristic feature of their secondary mass spectra is the fragmentation of their carboxyl groups, with the most significant neutral loss of H2O and CO2. For oxidized fatty acids with more double bonds, more oxygen-containing ring energy groups and higher overall unsaturation, the secondary fragmentation of arachidonic acid is significantly richer than that of arachidonic acid. The secondary fragments are often accompanied by bond breakage at the oxygen-containing functional groups and weak fatty acid chain breakage. The scope of their annotation can be expanded by making full use of their secondary information. At the same time, we summarized the LC-MS / MS structural fragmentation patterns of hydroxy-oxidized fatty acids reported in the literature, and included other fragmentation patterns of oxidized fatty acids, including the characteristic fragment of acetic acid at m / z59.0138 and the neutral loss of 46.0049Da. Epoxy isomers or other cyclic ketone lipids can also form m / z 59.0138, and the collision-induced dissociation of epoxy groups or cyclic ketone groups produces product ions with greater information content.

[0039] In-source cleavage of oxidized fatty acids: The structural types of oxidized fatty acids include hydroxyfatty acids, epoxy fatty acids and hydroperoxyl fatty acids. For oxidized fatty acids containing oxygen hydroxyl groups in their structure, especially those that can form conjugated double bond structures after fragmentation and dehydration, the in-source cleavage peak [MH-H2O]- can indicate the presence of hydroxyl groups in oxidized fatty acids, such as Figure 4 As shown in the arachidonic acid containing free hydroxyl groups and arachidonic acid containing free hydroxyl groups, the hydroxyl group can be broken and dehydrated at very low energy, indicating that the hydroxyl group on the chain is easier to fragment at low energy than the carboxyl group. Although the in-source fragmentation voltage is low, the in-source fragmentation is achieved because the hydroxyl group on the chain in the structure is easily broken and dehydrated [MH-H2O]- to form a new double bond and form a stable conjugated state with the existing double bond in the structure. For other oxidized fatty acids without free hydroxyl groups in the structure, there is no corresponding in-source water loss peak [MH-H2O]-. Previous studies often needed to avoid considering the in-source cleavage peak when qualitatively analyzing. In this method, we use the identification of the in-source cleavage peak to confirm that the structure is a polyunsaturated free hydroxyl oxidized fatty acid.

[0040] Construction of oxidized fatty acid mass spectral library: The self-built oxidized fatty acid mass spectral library includes the following sources, such as Figure 5 As shown in the figure, there are 209 secondary data of oxidized lipids reported in the literature, secondary data of 54 oxidized fatty acid standards collected in the early stage of this method, 47 spectra collected in GNPS, including 25 MH forms and 22 M-H2O-H forms, and 386 secondary spectra of oxidized lipid substances such as oxidized PC and PE oxidation products in public libraries such as MoNA. The information contained in the library includes the name, structure, predicted retention time and experimental secondary mass spectrum of oxidized fatty acids, the source and proportion of secondary mass spectrum information of oxidized fatty acid mass spectrum library Figure 5 .

[0041] Construction of the oxidized fatty acid structure library: 3441 oxidized fatty acids with a C number range of 10 to 30, a number of oxygen atoms range of 3 to 15, a total number of double bonds (unsaturation) range of 1 to 15, and a lipid category of "Fatty Acyls" were extracted from the LIPIDMAPS lipid structure library. Due to the highly isomerized nature of the molecules, only 241 unique molecular formulas correspond to the 3441 molecules. The chain length of the number of carbon atoms, the number of oxygen atoms, and the distribution of unsaturation are shown in the figure. Figure 6 As shown, the oxidized fatty acid structure library contains the oxidized fatty acid name, structural formula, molecular formula, accurate mass number, primary ion [MH]-, predicted secondary mass spectrum characteristic fragments obtained according to the summarized mass spectrum characteristic fragmentation rules, and retention time calculated according to the established retention time prediction model.

[0042] Construction of an in silico oxidized fatty acid extension library: The molecular formula forms not included in the oxidized fatty acid mass spectrum library and the oxidized fatty acid structure library are expanded within the range of 10 to 30 oxidized fatty acid carbon numbers, 1 to carbon chain length / 2 total number of double bonds, and 3 to carbon chain length / 2 oxygen numbers. The molecular formulas are supplemented and expanded. The database contains the molecular formula, molecular weight, accurate mass number, and corresponding mass-to-charge ratio [MH]-information of the corresponding oxidized fatty acids in negative ion mode, predicted secondary mass spectrum characteristic fragments obtained based on the summarized mass spectrum characteristic fragmentation rules, and retention time calculated based on the established retention time prediction model. The carbon chain length, oxygen number, and unsaturation distribution of the predicted molecular formula are as follows: Figure 7 shown.

[0043] Large-scale qualitative identification of oxidized fatty acids based on the self-built library of oxidized fatty acids: the liquid chromatography retention time, primary mass spectrum information and secondary mass spectrum information of the metabolites in the oxidized fatty acid extract obtained in the second step S2 are matched with the oxidized fatty acid mass spectrum library, oxidized fatty acid structure library and in silico oxidized fatty acid virtual structure library in S5 in turn; and the in-source fragmentation pattern [MH-H2O]-is extracted to infer whether the candidate compound is a hydroxyl-containing oxidized fatty acid; search condition 1 is that the mass window for matching the mass-to-charge ratio of the primary ion in the secondary mass spectrum obtained in S2 and the oxidized fatty acid mass spectrum library, oxidized fatty acid structure library, and in silico oxidized fatty acid virtual structure library is ±10ppm; search condition 2 is that the spectral entropy similarity between the secondary mass spectrum obtained in S2 and the secondary mass spectrum in the oxidized fatty acid mass spectrum library satisfies ≥0.6, and search condition 3 is that the secondary mass spectrum obtained in S2 and the oxidized fatty acid structure library, in silico The neutral loss and characteristic ions contained in the virtual structure library of silico oxidized fatty acids were matched with each other, and the mass window was ±10ppm (a successful match was considered as long as one of the four fragmentation modes was met); search condition 4 was that the retention time obtained in S2 matched the predicted retention time in the self-built library with a retention time matching window of ±1.5min; if the search results met one of the following three search conditions in sequence, they were all considered to be successfully identified, and if conditions 1, 2, and 4 were met, the target substance analyzed was a possible oxidized fatty acid, marked as Level 1; if conditions 1, 3, and 4 were met, the target substance analyzed was a possible oxidized fatty acid, marked as Level 2; if conditions 1, 2 or 3 were met, the target substance analyzed was a possible oxidized fatty acid, marked as Level 3; the rest of the cases were considered to be identification failures, indicating that the oxidized fatty acid did not exist in the sample; finally, by extracting the [M-H2O-H] form chromatographic peak at the same retention time as the corresponding candidate compound [MH] and the peak shapes of the two were consistent, it was determined whether the structure was likely to be a hydroxyl oxidized fatty acid, if so, it was a hydroxyl oxidized fatty acid, and if not, the structure did not contain fragile hydroxyl groups.

[0044] Taking the oxidized fatty acid with molecular formula C16H30O3 as an example, its specific qualitative process is explained as follows: Figure 8 As shown in Table 3, a total of 9 oxidized fatty acid isomers with a molecular weight of 270.2195 were obtained from the sample in the negative ion mode, such as Figure 8 A, respectively A(m / z 269.2122@t R 859.8), B(m / z 269.2122@t R 868.2), C(m / z269.2122@t R 874.8), D(m / z 269.2122@t R 885.0), E(m / z 269.2122@t R 899.4), F(m / z269.2122@t R 916.8), G(m / z 269.2122@t R 961.2), H(m / z 269.2122@t R 859.8), I(m / z269.2122@t R1004.4), and extracted the corresponding secondary mass spectrometry information from the original metabolomics data. Since no matching secondary mass spectra were found in the oxidized fatty acid mass spectrum database, the oxidized fatty acid structure library was continued to be matched and searched. In the negative ion mode, m / z 269.2122 could match 19 oxidized fatty acid candidates with the molecular formula of C16H30O3 in the oxidized fatty acid structure library, as shown in Table 2. The names were Ambrettolic acid, 16-hydroxy-9E-hexadecenoic acid, 16-hydroxy-9Z-hexadecenoic acid, (+)-12-hydroxy-9Z-hexadecenoic acid, 16-hydroxy-5-hexadecenoic acid, 16-hydroxy-6-hexadecenoic acid, 9S-hydroxy-7E-hexadecenoic acid, 2-keto palmitic acid, 3-keto palmitic acid, 4-keto palmitic acid, 5-keto palmitic acid, 7-keto palmitic acid, 8-keto palmitic acid, There are no reference spectra for these 19 metabolites, but since their molecular formulas are consistent, their mass deviations were calculated (experimental MS 1 -MS of molecules in the library 1 / Experimental MS 1 ), its MS 1 All of them were within 10 ppm (meeting search condition 1). The secondary of these 9 metabolites contained strong neutral loss characteristic fragments of H2O and CO2, m / z 251.20 and m / z 225.22 (meeting search condition 3). The predicted retention time of these 19 candidates in the library was 837.251±90s. The molecular A (t R :859.8), B(t R :868.2), C(t R :874.8), D(t R :885.0), E(t R :899.4), F(t R:916.8)(matches search condition 4), G(t R :935.4), H(t R :961.2), I(t R :1004.4) (does not meet search condition 4), the annotation results of the first 6 lipids are all FA 16:1; O, which corresponds to the 16 candidates of LIPID MAPS (as shown in Table 2), and the annotation level is Level 2. The annotation results of the last 3 lipids are all FA 16:1; O, and the annotation level is Level 3. By extracting its in-source fragmentation form m / z 251.2016 ([M-H2O-H]-), we found that only H ([MH]-, m / z 269.2122@859.8) The oxidized fatty acids have the corresponding in-source fragmentation forms H([MH-H2O]-, m / z 251.2017@859.8) , the peak shapes are similar, so it is speculated that there is a hydroxyl functional group in its structure and it is a hydroxy fatty acid.

[0045] Reference spectra of oxidized fatty acids, secondary spectra of identified oxidized fatty acids, and spectra of unknown metabolites are networked together for the analysis and annotation of unknown oxidized fatty acids: Through the study of actual samples, we found that due to the complexity of the matrix of the sample, some characteristic mass spectral fragments were not complete. Relying solely on characteristic fragment screening will miss the qualitative identification of very low oxidized fatty acids. However, due to the high heterogeneity of oxidized fatty acids in the sample, there are often structures with similar structures and very similar secondary mass spectra. Therefore, the annotation coverage can be greatly expanded through secondary similarity. Therefore, we networked all the spectra included in the mass spectral library of oxidized fatty acids, the secondary and unknown metabolic spectra of the identified fatty acids, and analyzed the unknown metabolites. The structures of unknown metabolites are inferred by means of biotransformation and modification reactions.

[0046] like Fig. 9 As shown in the figure, FA 18:2+3O (m / z 327.217@tR 622) is the secondary mass spectrum (red node) included in the library. We can associate it with the four oxidized fatty acids that have been identified in the sample from right to left: A (309.207@t R 725), B(309.207@t R 624), C(309.207@t R 784), D(309.207@t R 745), according to the conventional metabolic reactions, the annotation of the edge searches for the conventional modification type based on the difference in m / z of the parent ions corresponding to the secondary mass spectra represented by the two nodes, and can match the metabolic reaction types such as H2, O and H2O. By extending the annotation, it can be associated with the secondary E(325.202@t R 646) and known node D(309.207@tR 745) The secondary similarity is greater than the threshold. Although no characteristic fragments were found in the secondary spectrum of E, this point can be annotated as FA18:3+3O based on the structural similarity and conventional metabolic reaction types.

[0047] Application We applied the above strategy to the large-scale qualitative analysis of oxidized fatty acids in rice stems, and obtained 117 oxidized fatty acids, including 6 Level 1, 105 Level 2, and 6 Level 3 oxidized fatty acids, as shown in Table 3.

[0048]

[0049]

[0050] Table 1 63 common metabolic reactions and modification types

[0051]

[0052]

[0053] Table 2 Oxidized fatty acid standard sample information

[0054]

[0055] Table 3 Candidate oxidized fatty acids with molecular formula C16H30O3 m / z 269.2122 found in actual samples

[0056]

[0057]

[0058]

[0059]

[0060] Table 4 Preliminary qualitative results of oxidized lipids in actual samples

Claims

1. A method for the large-scale qualitative characterization of oxidized fatty acids based on non-targeted metabolomics, characterized by: S1, pre-treating the biological sample to be tested and extracting oxidized fatty acid components in the sample to be tested; S2, using ultra-high performance liquid chromatography-high resolution mass spectrometry to perform non-targeted metabolome analysis on the pre-treated oxidized fatty acid extract, collecting the liquid chromatography retention time of the metabolites, the primary mass spectrometry information including the mass-to-charge ratio and intensity information of the primary ion, and the secondary mass spectrometry information including the mass-to-charge ratio and intensity information of the secondary ion; the primary ion here refers to the ion directly collected after the compound is ionized; Secondary ions refer to the ions collected after a compound is ionized and then collided and fragmented with a certain amount of energy; S3, by purchasing actual oxidized fatty acid standards (such as 12-hydroxydodecanoic acid, 3-hydroxydecanoic acid, Prostaglandin D2, (+ / -)12(13)-EpOME, 9-HODE, etc., covering chemical structures with carbon atoms between 10-30, oxygen atoms between 2-15, and double bonds between 2-15 and not less than 50 (more than 50), and obtaining their corresponding chromatographic retention times, constructing a structural retention time relationship and regression equation of the carbon number, total number of double bonds, and oxygen number in the oxidized fatty acid molecular formula relative to the retention time: predicted retention time = k1* carbon number of fatty acid + k2* unsaturation + k3* oxygen number, k is the obtained model parameter; S4, the difference between the primary parent ion and the secondary fragment ion is called neutral loss. The secondary fragmentation patterns of oxidized fatty acids in negative ion mode are: 1) neutral loss of carbon dioxide (CO2), 2) neutral loss of water (H2O) and carbon dioxide (CO2), 3) m / z 59.0138 characteristic fragment of acetic acid and neutral loss of 46.0049Da, 4) neutral loss of acetic acid at 46.0049Da and neutral loss of water, satisfying one or more of the four conditions; S5, constructing a self-built database of oxidized fatty acid ultra-high performance liquid chromatography-high resolution mass spectrometry, which includes three sub-databases: oxidized fatty acid mass spectrum library, oxidized fatty acid structure library, and in silico oxidized fatty acid virtual structure library. The specific construction method is as follows; 1) Construction of an oxidized fatty acid mass spectrum library: Collect oxidized fatty acid information from mass spectrum databases (such as the Global Natural Products Social Molecular Networking (GNPS) mass spectrum database) and literature reports (such as the secondary information of oxidized fatty acids disclosed in Cell Chem Biol. 2019, 26(3): 433-442) to construct an oxidized fatty acid mass spectrum library. The database includes the following information: molecular information of oxidized fatty acids collected in the mass spectrum library or / and literature, including the name of the oxidized fatty acid, molecular structure, molecular formula, accurate mass number, primary parent ion [MH]-, secondary mass spectrum information, and retention time predicted by the model in S3; 2) Construction of the oxidized fatty acid structure library: Collect all oxidized fatty acids in the lipid structure database (Lipid Metabolites And Pathways Strategy (LIPDMAPS)) whose molecular formulas contain 10-30 carbon atoms, 3-15 oxygen atoms, and 1-15 double bonds for all lipids of the "Fatty Acyls" category to construct an oxidized fatty acid structure library. The database includes the oxidized fatty acid name, molecular structure formula, molecular formula, accurate mass number, primary ion [MH]-, secondary mass spectrum information, and retention time predicted according to the model in S3; 3) Construction of in silico oxidized fatty acid virtual structure library: For molecular formula forms not present in 1) and 2), the molecular formulas were supplemented and expanded within the range of 10 to 30 oxidized fatty acid carbon numbers, 1 to carbon chain length / 2 total number of double bonds, and 3 to carbon chain length / 2 oxygen number, and their corresponding molecular formulas, molecular weights, accurate mass numbers, and corresponding mass-to-charge ratios [MH]-information in negative ion mode, secondary mass spectrometry characteristic fragments predicted in S4, and retention times predicted by the model in S3 were obtained; S6, matching the liquid chromatography retention time, primary mass spectrometry information and secondary mass spectrometry information of the metabolites in the oxidized fatty acid extract obtained in the second step S2 with the oxidized fatty acid mass spectrum library, oxidized fatty acid structure library and in silico oxidized fatty acid virtual structure library in S5 in sequence; and inferring whether the candidate compound is a hydroxyl-containing oxidized fatty acid by extracting the in-source fragmentation pattern [MH-H2O]-; S7, all the secondary mass spectra of oxidized fatty acids collected in the oxidized fatty acid mass spectral library in S5, the secondary mass spectra of oxidized fatty acids collected in S2 and successfully matched in S4, and the secondary mass spectra of all unknown metabolites collected in S2 and not successfully matched in S4 were co-constructed into a molecular network on the GNPS platform based on the similarity of secondary mass spectra, and the unknown oxidized fatty acid molecules that were successfully associated with the molecular network were extendedly annotated with the oxidized fatty acids and their modified metabolites through the collected 63 common metabolic reactions and modification types (see Appendix 1).

2. The method according to claim 1, characterized in that: In S1, in order to cover as many oxidized fatty acid components as possible, an antioxidant (butylated hydroxytoluene, with a final concentration of 0.4-0.6 mM in the extract) is added to the extract of the biological sample to be tested (1 mL of water) to prevent the chemical properties of the oxidized fatty acid components in the sample from changing, and the oxidized fatty acid components in the sample are enriched by liquid-liquid extraction (the extractant includes 250-350 μL of water and 400-500 μL of ethyl acetate, mixed and layered, the upper layer is taken, and extracted twice (400-500 μL of ethyl acetate is added to the lower layer for the second time, and the upper layer is taken), and the extracts are combined), and a metabolomics pretreatment method is used, including freeze-drying, re-dissolving (the re-dissolving solvent is 180-220 μL of methanol solution containing 0.45% to 0.55% formic acid by volume) and centrifugation to remove the precipitate of the biological sample to be tested.

3. The method according to claim 1, characterized in that: In S2, ultra-high performance liquid chromatography-high resolution mass spectrometry was used for analysis, and the mass spectrometry scanning mode was full scan plus secondary mass spectrometry acquisition mode full MS / data-dependent ddMS2.

4. The method according to claim 1, characterized in that: In S6, the liquid chromatography retention time, primary mass spectrum information and secondary mass spectrum information of the analyte obtained in the second step are searched in the oxidized fatty acid mass spectrum library, the oxidized fatty acid structure library and the insilico oxidized fatty acid virtual structure library in sequence; if the search results meet one of the following three search conditions in sequence, it indicates that the identification is successful, and if conditions 1, 2, and 4 are met, the target substance analyzed is a possible oxidized fatty acid, which is marked as Level 1; if conditions 1, 3, and 4 are met, the target substance analyzed is a possible oxidized fatty acid, which is marked as Level 2; if conditions 1, 2 or 3 are met, the target substance analyzed is a possible oxidized fatty acid, which is marked as Level 3; the rest of the cases are considered as identification failures, indicating that the oxidized fatty acid does not exist in the sample; Search condition 1: The mass window for matching the mass-to-charge ratio of the primary ions in the secondary mass spectra obtained in S2 and the oxidized fatty acid mass spectrum library, the oxidized fatty acid structure library, and the in silico oxidized fatty acid virtual structure library was ±10 ppm; Search condition 2: The spectral entropy similarity between the secondary mass spectrum obtained in S2 and the secondary mass spectrum in the oxidized fatty acid mass spectrum library satisfies ≥ 0.6, Search condition 3: The secondary mass spectrum obtained in S2 was matched with the neutral loss and characteristic ions contained in the oxidized fatty acid structure library and the in silico oxidized fatty acid virtual structure library, with a mass window of ±10 ppm (a successful match was considered if one of the four fragmentation patterns was met); Search condition 4: the retention time matching window between the retention time obtained in S2 and the predicted retention time in the self-built library was ±1.5 min; Finally, by extracting the [M-H2O-H] form chromatographic peak at the same retention time of the corresponding candidate compound [MH] and the peak shapes of the two are consistent, it is judged whether the structure is likely to be a hydroxyl-oxidized fatty acid. If so, it is a hydroxyl-oxidized fatty acid. If not, the structure does not contain fragile hydroxyl groups.

5. The method according to claim 1, characterized in that: In S5, the secondary information of all oxidized fatty acids included in the self-built library from the mass spectrometry database and the secondary mass spectra of all oxidized fatty acids identified in S6 are used to construct a molecular network based on the secondary mass spectra similarity with the secondary mass spectra obtained in S2. The secondary mass spectra similarity is cosine similarity, and the similarity of the two spectra for network construction must be greater than 0.6, with a minimum of 5 fragments. For the two successfully associated metabolites, the mass difference between the two metabolites is matched with the mass numbers of conventional metabolic reactions and modification types in a mass window of ±10ppm, and the structure of the successfully matched unknown oxidized fatty acid molecules is inferred (based on the molecular weight of the two associated nodes plus or minus a matched modification group), mainly for annotation of oxidized fatty acid derivatives and modifications, and the successfully annotated compounds are marked as Level 3.

6. The method according to claim 1 or 2, characterized in that: The biological sample to be tested is one or more of human, animal, and plant samples.

Images