Method and system for determining similarity by comparing 1H-NMR spectra with 1H-1H COSY NMR spectra

Abstract

The present invention relates to techniques for estimating molecular structures, and provides a method for calculating the similarity between two spectrograms by using not only 1DNMR spectrogram information but also 2D NMR spectrogram information, in which spectrograms of an estimated target material are compared with spectrograms of candidate materials, determining similar candidate molecular structures, and calculating the similarity between the two spectrograms using not only 1DNMR spectrogram information but also 2D NMR spectrogram information. Therefore, the molecular structure of the estimated target material is estimated. Further, when calculating the similarity, an accurate molecular structure of a complex aromatic material can be provided.

Description

Technical Field

[0001] The present invention relates to a technology that enables researchers to effectively estimate the molecular structure of an unknown substance by calculating a plurality of candidate molecular structures and excluding inappropriate candidate molecular structures from the plurality of candidate molecular structures to reduce the number of candidate molecular structures.

[0002] Furthermore, the present invention relates to 1 H-NMR spectrum analysis and comparison is a technique for extracting final candidate molecular structures from a reduced number of candidate molecular structures. Background Art

[0003] 1H-NMR spectrum analysis is performed to estimate the molecular structure of the unknown substance. At this time, the analyst obtains all possible candidate molecular structures from the obtained 1H-NMR spectrum and predicts the peak positions and shapes of all hydrogens in each candidate molecular structure to create a virtual spectrum. Thereafter, the most reasonable structure is selected by comparing the virtual spectrum with the actual spectrum (in terms of symbols, in the present invention, 1H-NMR and 1 H-NMR, 1H-1H NMR and 1 H- 1 H COSY NMR were considered identical).

[0004] However, when a large number of peaks are concentrated in a small area, as in the case of aromatic macromolecules, there is a problem that the above method is difficult to use for comparing similar molecules.

[0005] For example, the following prior art does not use information such as peak multiplicity in the spectrum and thus cannot reflect detailed structural information, which results in an inability to derive accurate molecular structure estimation results.

[0006] A Structure Elucidation System Based on 1 H NMR and HH COSY Spectra inOrganic Chemistry,Hideyuki Masui and Huixiao Hong,J.Chem.Inf.Model.2006,46,2, Summary of the invention

[0007] Technical issues

[0008] The present invention aims to solve the problems of the prior art and to provide a method and system capable of performing accurate 1H-NMR spectrum analysis even for a substance having a large molecular size or a complex shape, such as an aromatic ring.

[0009] Technical Solution

[0010] In order to solve the above technical problems, the present invention provides a method for predicting the molecular structure of a target substance from the 1H-NMR spectrum of the target substance, wherein the method comprises: a key spectrum acquisition step for obtaining the 1H-NMR spectrum of the target substance; a candidate spectrum information acquisition step for obtaining spectrum information of a virtual spectrum of each candidate substance from the candidate molecular structure of the target substance as candidate spectrum information; a spectrum comparison step for comparing the key spectrum information and the candidate spectrum information to calculate the similarity between the key spectrum and the candidate spectrum; and a target substance molecular structure determination step for obtaining the candidate spectrum with the highest similarity in the spectrum comparison step and selecting the molecular structure of the corresponding candidate substance as the molecular structure of the target substance.

[0011] In addition, the present invention provides a method for determining similarity by comparing two NMR spectra, wherein the method includes: a peak information extraction step of extracting peak information for spectrum comparison from each peak of a first spectrum as a key spectrum and a second spectrum as a candidate spectrum; a peak correspondence step of matching each peak of the first spectrum and the second spectrum; and a spectrum similarity calculation step of calculating the similarity between the first spectrum and the second spectrum by comparing the peak information of the peaks of the matched first spectrum and the second spectrum.

[0012] In addition, the present invention provides a system for estimating the molecular structure of a target substance, wherein the system includes: a spectrum acquisition device that acquires a 1H-NMR spectrum and a 1H-1H NMR spectrum of the target substance from an NMR device as a key spectrum of the target substance; a candidate molecular structure acquisition module that acquires candidate molecular structures of the target substance and calculates a candidate molecular structure list; a virtual spectrum calculation module that calculates a virtual spectrum from the candidate molecular structures included in the candidate molecular structure list as a candidate spectrum; a spectrum information extraction module that extracts spectrum information from each of the key spectrum and the candidate spectrum; a spectrum similarity calculation module that calculates the similarity between the key spectrum and the candidate spectrum by using the spectrum information of the key spectrum and the candidate spectrum; and a final estimated structure output module that outputs the candidate molecular structure corresponding to the candidate spectrum with the highest similarity as the final estimated molecular structure of the target substance.

[0013] In addition, the present invention extracts hydrogen structural features for identifying hydrogen with predetermined structural features from candidate molecular structures, and verifies the suitability of the candidate molecular structures (which includes the step of verifying whether all hydrogen with predetermined structural features can be assigned to all peaks of the 1H-NMR spectrum), verifies each candidate molecular structure to update the candidate molecular structure list, excludes inappropriate candidate molecular structures, thereby reducing the number of candidate molecular structures to improve computational efficiency.

[0014] To this end, the system for estimating the molecular structure of a target substance according to the present invention includes: a hydrogen structure feature recognition module for identifying hydrogen atoms with predetermined structural features in each candidate molecular structure of the target substance; a peak allocation validity determination module, which verifies whether each hydrogen of the candidate molecular structure can be assigned to all peaks in the spectrum based on the structural feature information of each hydrogen identified from the candidate molecular structure; and a valid candidate molecular structure output module for updating the candidate molecular structure list by using the result of the peak allocation validity judgment module.

[0015] Beneficial Effects

[0016] The present invention provides a method for calculating the similarity of two spectra by using 2D NMR spectrum information and 1D NMR spectrum information, so as to bring about the effect of accurately and effectively estimating the molecular structure of a target substance by comparing the spectrum of the target substance with the spectrum of a candidate substance.

[0017] Furthermore, the present invention quickly excludes inappropriate candidate substances from the candidate molecular structures, and thus brings about the effect of more quickly estimating the accurate molecular structure of the unknown substance through similarity calculation. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The following drawings attached herein illustrate preferred embodiments of the present invention by way of example, and together with the above detailed description of the present invention, are used to enable a further understanding of the technical concept of the present invention, and therefore, the present invention should not be construed as being limited to the matters described in such drawings.

[0019] Figure 1 is an example of the COSY spectrum used in the present invention.

[0020] Figure 2 The flowchart shows the steps for calculating the similarity of spectra and estimating the molecular structure according to the present invention.

[0021] Figure 3 A block diagram of a system for calculating spectral similarity and estimating molecular structure according to the present invention. DETAILED DESCRIPTION

[0022] The present invention uses a 1H-1H COSY spectrum as one of the 1H-NMR spectrum and the 2D NMR spectrum of a substance. The 1H-1H COSY spectrum is obtained by a nuclear magnetic resonance spectrometer (NMR). Both the horizontal axis and the vertical axis are composed of the 1H spectrum in the 1H-1H COSY spectrum, and Figure 1 In the present invention, 1H-NMR spectrum, 1 H-NMR spectrum, 1H-1H COSY NMR spectrum and 1 H- 1The H COSY NMR spectra each show the same thing.

[0023] 1. Method for estimating the molecular structure of a target substance according to the present invention

[0024] The method for estimating the molecular structure of a target substance according to the present invention is an invention that obtains spectrum information (this is called a virtual spectrum) corresponding to the molecular structures of a plurality of candidate substances, and then compares the virtual spectrum information of the candidate substances with the spectrum information of the target substance (this is called a key spectrum) to select the virtual spectrum that is most similar to the key spectrum, and selects the molecular structure of the corresponding candidate substance as the molecular structure of the target substance.

[0025] The steps according to the present invention mainly consist of the following three steps. First, A. a step of obtaining the spectrum information of the target substance to be estimated molecular structure, followed by B. a step of selecting the candidate molecular structure of the target substance, and then C. a step of comparing the virtual spectrum of the candidate molecular structure with the spectrum of the target substance to obtain the final molecular structure of the target substance from the candidate molecular structure. Figure 2 Each step is described in more detail.

[0026] A. Steps to obtain the spectral information of the target substance

[0027] (1) Key spectrum acquisition step S10

[0028] This is the step of obtaining the 1H-NMR spectrum and 1H-1H COSY spectrum of the target substance by NMR. The obtained 1H-NMR spectrum and 1H-1H COSY spectrum are called key spectra.

[0029] (2) Key spectrum information acquisition step S20

[0030] Get spectrum information from key spectrum. Spectrum information includes the following information.

[0031] (a) Peak position (chemical shift value): the x-axis value of each peak in the one-dimensional spectrum

[0032] (b) Peak splitting value (multiplicity (split)): The number of small peaks that form each peak (e.g. Figure 1 In the spectrum of , the splitting of the leftmost peak is 2, and the splitting of the rightmost peak is 3)

[0033] (c) Peak integral value: the area value of each peak

[0034] (d) COSY information (COSY): Information of all peak pairs in which COSY peaks are observed.

[0035] Example) Figure 1In the two-dimensional key spectrum of , a red dot is shown at the peak position associated with the first peak on the left, and this is called a COSY peak. Since the first peak and the seventh and eighth peaks show COSY peaks, for the matrix whose rows and columns are peak lists, the elements at the positions (1, 7) and (1, 8) are set to 1.

[0036] Figure 1 is a diagram showing a predetermined two-dimensional key spectrum (a) and a COSY matrix showing information of its COSY peaks (b), and in Figure 1 In (a), if there is a COSY peak between the two 1D peaks, Figure 1 The element value of the COSY matrix in (b) is set to 1. In the above example, since the first peak (1) shows a COSY peak together with the seventh peak (7) and the eighth peak (8), it can be seen that Figure 1 The element values ​​of COSY(1,7) and COSY(1,8) in the COSY matrix of (b) are each set to 1.

[0037] B. Steps for selecting candidate molecular structures of target substances

[0038] The steps of selecting a candidate molecular structure of a target substance will be described.

[0039] (1) Step S30 of obtaining candidate molecular structures

[0040] This is the step of obtaining candidate molecular structure information of the target substance. The candidate molecular structure is a molecular structure of a candidate that the target substance may have, so as to finally estimate what molecular structure the target substance has. By using the measurement information of the target substance, such as the mass spectrum of the target substance, information of a plurality of candidate molecular structures that may be the molecular structure of the target substance can be estimated.

[0041] A plurality of candidate molecular structures are obtained by an analysis method such as LC-MS / MS, and for example, the candidate molecular structures can be obtained by obtaining the molecular weight and molecular formula of the target substance and then searching for a molecular structure with the same molecular formula in a known substance database (SCIFinder, PubChem, etc.). Another example is a method of creating possible candidate molecular structures by obtaining partial structural information of a molecule from an MS / MS analysis result and then combining the partial structural information.

[0042] A candidate molecular structure list including a plurality of candidate molecular structures of the target substance acquired as described above is created.

[0043] (2) Step S40 of testing the validity of the candidate molecular structure

[0044] According to the step of obtaining the final molecular structure of the target substance from the candidate molecular structures (described in step C below), the molecular structure of the target substance can be obtained by comparing the spectra of the candidate molecular structures of the target substance obtained above.

[0045] However, in this case, spectra of many candidate molecular structures should be compared, so that in order to reduce the number of candidate molecular structures, a step for verifying the suitability of the candidate molecular structures may be performed before the spectrum comparison step.

[0046] The candidate molecular structure validity step extracts hydrogen feature information from a plurality of candidate molecular structures, determines the validity of unsuitable candidate molecular structures based on whether all hydrogens of each candidate molecular structure can be assigned to all peaks in the spectrum of the target substance, and excludes unsuitable candidate molecular structures from the list of candidate molecular structures. This step follows the detailed steps below.

[0047] (2-1) Hydrogen structure feature information extraction step S41

[0048] This is a step of identifying the structural characteristic information of hydrogen of each candidate molecular structure included in the list of candidate molecular structures to extract the hydrogen characteristic information of each candidate molecular structure. The hydrogen characteristic information to be identified includes the following:

[0049] ①The number of isostructural hydrogen

[0050] This refers to the number of hydrogen atoms whose bonding structure is structurally isostructural to the bonding structure of the target hydrogen atom. The number of isostructural hydrogen atoms becomes the integral value of the corresponding hydrogen peak.

[0051] ②The number of hydrogen atoms connected to the first neighboring atom

[0052] It is the number of hydrogens bonded to the first neighboring atom in the atom bonded to the target hydrogen. The multiplicity of the target hydrogen is determined by the number of hydrogens bonded to the first neighboring atom bonded to the first neighboring atom.

[0053] ③The number of hydrogen atoms connected to the second adjacent atom

[0054] It is the number of hydrogens connected to the second connected atom via the first neighbor atom among the atoms connected to the target hydrogen. That is, it is the number of hydrogens connected to the atom directly connected to the first neighbor atom. This affects the multiplicity of the target hydrogen and is used to predict the shape of the peak.

[0055] ④Other hydrogens that may have COSY peaks

[0056] Identify other hydrogens that may have the same COSY peak as the target hydrogen. If the distance between the two hydrogens is within a certain bonding distance, the COSY signal is displayed.

[0057] Therefore, by extracting structural characteristic information of each hydrogen from each candidate molecular structure, it is determined whether each hydrogen of each candidate molecular structure can be assigned to each peak of the key spectrum of the target substance based on the characteristic information of the hydrogen of each candidate molecular structure in the next step.

[0058] (2-2) Peak Allocation Possibility Determination Step S42

[0059] This is the step that determines whether all hydrogens of a given candidate molecular structure can be assigned to all peaks of the key spectrum.

[0060] If all hydrogens of a given candidate molecular structure cannot be assigned to all peaks in the key spectrum, the molecular structure validity test according to the present invention determines that the corresponding candidate molecular structure is an invalid molecular structure.

[0061] Therefore, in order to determine whether an assignment is impossible, all possible assignment methods should be checked, which requires a very large number of calculations. For example, if the predetermined candidate molecular structure contains m distinguishable hydrogens and there are n peaks in the spectrum, there are mPn possible assignment methods, so that the time to determine the suitability of the assignment increases exponentially according to the complexity of the molecule.

[0062] Therefore, the present invention determines the probability of assignment by replacing the problem of assigning all hydrogens of a candidate molecular structure to each peak of a key spectrum with a mixed integer linear programming (MIP) problem with linear constraints.

[0063] Here, since the MIP problem has no optimization objective, slack variables s and t are introduced, and the probability of allocation is determined based on the success / failure of the optimization itself.

[0064] The formula used for optimization is the following formula 1, and the optimization is to obtain the optimal solutions si and ti that minimize formula 1.

[0065] [Formula 1]

[0066] For a matrix X of size M×N

[0067] make minimize

[0068] (If hydrogen j is not assigned to peak i, then X ij = 0, if hydrogen j is assigned to peak i, then X ij =1,j=1,2,…,M,i=1,2,…,N)

[0069] <Constraints>

[0070] The constraints used for the calculation of Equation 1 are as follows.

[0071] ①For 1≤i≤N, s i ≥0,t i ≥0

[0072] ②For all 1≤i≤N, the peak integral value of hydrogen j is intg 氢 (i), and the integral value of peak i of the key spectrum is intg 峰 (i)

[0073]

[0074]

[0075] ③X ij = 0, (if the multiplicity of hydrogen j is different from the multiplicity of peak i)

[0076] ④X ij = 0, (if the peak integral value of hydrogen j is greater than the peak integral value of peak i)

[0077] ⑤ For two peaks i_1 and i_2 having COSY peaks, there must be at least one COSY peak among the set of all hydrogens C_hydrogen(i1) assigned to i_1 and the set of all hydrogens C_hydrogen(i2) assigned to i_2.

[0078] (2-3) Candidate Molecular Structure List Update Step S43

[0079] In the optimized formula 1, if the optimization is successful and the optimal solution of s and t is obtained, all hydrogens of the corresponding structure can be assigned to the peaks of the key spectrum, and this is determined to be a valid structure. If the optimization fails and the optimal solution is not obtained, it is considered that there are one or more unassignable hydrogens, so that the corresponding candidate structure is determined to be a candidate molecular structure that is not suitable for the given spectrum.

[0080] Therefore, in this step, unsuitable candidate molecular structures are excluded, and the candidate molecular structure list is updated to output a list of valid candidate molecular structures.

[0081] C. Final Molecular Structure Selection Step

[0082] This is a step of comparing the spectrum information of the candidate molecular structures included in the obtained candidate molecular structure list or the effective candidate molecular structure list with the spectrum information of the target substance to select the final molecular structure of the target substance. This is achieved through the following detailed steps.

[0083] (1) Candidate spectrum information acquisition step S50

[0084] A virtual spectrum corresponding to each candidate molecular structure is obtained from the information of the plurality of candidate molecular structures obtained above (which is set as the spectrum of each candidate molecular structure), and the information of each spectrum is obtained. When the molecular structure information is input, a known virtual spectrum information calculation device or software for calculating virtual spectrum information is used to obtain a virtual spectrum of each candidate molecular structure.

[0085] The candidate spectrum information, which is spectrum information of candidate spectra of the candidate molecular structure to be obtained, includes items such as the above-mentioned key spectrum information.

[0086] (2) Spectrum comparison step S60

[0087] By comparing the spectrum information (each peak data) of the key spectrum (which is the spectrum of the target substance) with the spectrum information of each candidate, the similarity between the key spectrum and the candidate spectrum is calculated. The spectrum comparison method according to the present invention to be described below is applied to spectrum comparison.

[0088] The virtual spectrum of the candidate substance that is most similar to the key spectrum is obtained according to the spectrum similarity calculation steps to be described below.

[0089] The steps of determining the similarity by comparing the NMR spectra of two molecules according to the present invention will be described.

[0090] (2-1) Extracting spectrum comparison information from key spectra and candidate spectra S61

[0091] The comparison information used for spectrum comparison is extracted from each peak of the key spectrum and the candidate spectrum. The comparison information is the above-mentioned spectrum information.

[0092] (2-2) Peak correspondence step S62 of matching the peak of the candidate spectrum with the peak of the key spectrum

[0093] This is the step of assigning peak Qi of the candidate spectrum to peak Ki of the key spectrum. Here, i is the number of peaks in the key spectrum and the candidate spectrum.

[0094] When matching Qi with Ki, the following rules should be satisfied.

[0095] ①The integral value of Qi should not be greater than the integral value of Ki.

[0096] ②The split value of Qi should be the same as the split value of Ki.

[0097] ③ One or more hydrogens with COSY relationship with Qi should be assigned to other peaks with COSY relationship with Ki.

[0098] The COSY peak assignments for the query peak Qi and key peak Ki will refer to Figure 1 (c) and (d) are described.

[0099] If the cosy matrix of the key spectrum and the cosy matrix of the query spectrum are Figure 1 As shown in (c) and (d), the constraint means that when there is a COSY peak among the key peaks i1 and i2, there must be at least one mutual COSY peak among the query peak set C_query(i1) assigned to i1 and the query peak set C_query(i2) assigned to i2.

[0100] Considering that K11 and K13 are both assigned to the Q7 peak, and K8, K9 and K10 are all assigned to the Q8 peak, Figure 1 In (c), the Q7 peak and the Q8 peak have COSY peaks (matrix elements are 1), so that there should be a COSY peak among the peaks assigned to Q7 and the peaks assigned to Q8. Figure 1 In (d), there is no COSY peak between the K11 and K13 groups and the K8, K9 and K10 groups, and such an assignment is an inappropriate assignment.

[0101] (2-3) Similarity calculation step of two spectra S63

[0102] According to the above steps, the peak Qi (i=1, 2, ..., M) of the candidate spectrum (called the query peak) is assigned to the peak Kj (j=1, 2, ..., N) of the key spectrum (called the key peak), and then the similarity between the candidate spectrum and the key spectrum is calculated to obtain the best similar candidate spectrum with the highest similarity.

[0103] The similarity is calculated by the objective function of the following formula 2, and the best similar candidate spectrum with the highest similarity is the best similar candidate spectrum that reduces the objective function value in formula 2.

[0104] [Formula 2]

[0105]

[0106] (M and N are the number of peaks in the query spectrum and the key spectrum, respectively, X is an M×N binary matrix, and if the query peak Qi is not assigned to the key peak Kj, then X ij = 0, and if the query peak Qi is assigned to the key peak Kj, then X ij = 1. Displacement 查询 (i) and displacement 关键 (j) represent the chemical shifts of Qi and Ki respectively. )

[0107] At the same time, the calculation of the objective function has the following restrictions.

[0108] ① (intg 查询 (i) is the integral value of the query peak Qi, and intg关键 (j) is the integral value of the key peak Kj)

[0109] ②If the multiplicity of Kj is different from the multiplicity of Qi, then X ij =0

[0110] ③If the integral value of Qi is greater than the integral value of Kj, then X ij =0

[0111] ④ For two key peaks Ki_1 and Ki_2 with COSY peaks, there must be at least one COSY peak in all query peak sets C_query(Ki_1) assigned to Ki_1 and all query peak sets C_query(Ki_2) assigned to Ki_2.

[0112] Therefore, while applying the above constraints, the similarity between two spectrograms is calculated by calculating the objective function according to Equation 2 above.

[0113] (3) Target substance molecular structure determination step S70

[0114] The molecular structure of the candidate substance corresponding to the candidate spectrum obtained as the most similar in the spectrum similarity calculation step is determined as the molecular structure of the target substance.

[0115] (3-1) Final Candidate Molecular Structure Determination Step S71

[0116] The candidate spectrum with the minimum objective function calculation value to be described below is determined as the best similarity spectrum with the highest similarity. There is a final candidate molecular structure determination step, in which the candidate structure corresponding to the best similarity spectrum is estimated as the molecular structure of the target substance. The determined final candidate molecular structure is determined as the molecular structure of the target substance to be estimated.

[0117] (3-2) Deuterium Substitution Position Estimation Step S72

[0118] The present invention may further include a deuterium substitution position estimating step of estimating the deuterium substitution position by spectrum comparison.

[0119] If some hydrogens of a specific molecule are substituted with deuterium, the 1H NMR peak of the corresponding hydrogen has a unique characteristic form, and if the deuterium substitution rate is not 100%, the corresponding peak appears in the form of a small peak having an integral value of a decimal point unit instead of an integer, and the corresponding peak appears in the form of a mixture of a single peak and multiple peaks. That is, the 1H-NMR peak at the deuterium substitution position has a characteristic shape.

[0120] At the same time, it is known which hydrogen of the candidate molecular structure corresponds to each peak Qi of the candidate spectrum estimated from the candidate molecular structure, and since each Qi is assigned to each Kj in the above assignment process, the hydrogen assigned to each Kj can be known. Therefore, since the Qi assigned to the deuterium substitution peak with a characteristic shape can be known, it is known which hydrogen of the candidate molecular structure is ultimately replaced by deuterium. When a deuterium substitution reaction is performed, the deuterium substitution position can be used to confirm whether the substitution occurs at the desired position.

[0121] By applying the above principles, the deuterium substitution position estimation step finds the peak of a mixture of single peaks and multiple peaks in the peak Kj of the key spectrum, finds the peak of Qi corresponding to the best similarity spectrum, and finds the corresponding hydrogen in the final estimated candidate molecular structure to identify the position of the hydrogen substituted by deuterium in the molecular structure of the target substance.

[0122] 2. System for estimating molecular structure of target substance according to the present invention

[0123] In the following, reference is made to Figure 3 , a molecular structure estimation system for estimating the molecular structure of a target substance according to the present invention will be described.

[0124] (1) Spectrum acquisition device 10

[0125] This is a composition of obtaining the 1H-NMR spectrum and 1H-1H NMR spectrum of the target substance from the NMR device. It can be composed of an input unit of a computer input device that receives spectrum information from a typical NMR device. In this case, the spectrum of the target substance is called a key spectrum.

[0126] (2) Spectral information extraction module 20

[0127] It is a composition for obtaining spectrogram information from a spectrogram. The obtained spectrogram information is the same as described in the spectrogram information obtaining step. The spectrogram information extraction module obtains key spectrogram information including information of key spectrogram peak Ki from the key spectrogram, and obtains candidate spectrogram information including information of candidate spectrogram peak Qi of the candidate spectrogram.

[0128] (3) Candidate molecular structure acquisition module 30

[0129] It is to receive the result of mass spectrometry of the target substance, for example, and estimate and calculate the composition of the candidate molecular structure of the target substance. As mentioned above, the candidate molecular structure can be obtained by a method such as mass spectrometry.

[0130] In the present invention, a candidate molecular structure that may be the molecular structure of the target substance is obtained using a known method, so that the candidate molecular structure acquisition module of the present invention can obtain the molecular weight and molecular formula of the target substance from the LCMS / MS analyzer, and search the molecular weight and molecular formula in a known database to obtain the candidate molecular structure, or can be composed of a data input module that receives the candidate molecular structure obtained by the known method as described above.

[0131] (4) Virtual spectrum calculation module 40

[0132] The virtual spectrum calculation module calculates the virtual spectrum of each candidate molecular structure of the target substance calculated by the candidate molecular structure acquisition module. The spectrum of each candidate molecular structure calculated by the virtual spectrum calculation module is input to the spectrum information extraction module 20 to extract the candidate spectrum information. The virtual spectrum calculation module 40 is composed of a known virtual spectrum information calculation device or software that calculates virtual spectrum information when molecular structure information is input.

[0133] (5) Spectral similarity calculation module 50

[0134] The spectrum similarity calculation module matches the peak Qi of the candidate spectrum calculated by the spectrum information extraction module 20 with the peak Kj of the key spectrum, and calculates the similarity between the key spectrum and the candidate spectrum.

[0135] The matching method of Kj and Qi and the method for calculating the spectrum similarity are the same as those described in 2.(2) and 2.(3) above. It is the spectrum with the highest similarity to the candidate spectrum, for which the objective function of Formula 1 is calculated to be the minimum.

[0136] (6) Final estimated structure output module 60

[0137] The candidate structure output module determines the candidate molecular structure corresponding to the corresponding candidate spectrum from the best similarity spectrum calculated by the above spectrum similarity calculation module 50 as the final estimated structure of the target substance, and outputs the corresponding molecular structure information.

[0138] (7) Deuterium substitution position calculation module 70

[0139] This is a module that determines a candidate molecular structure having the best similarity spectrum calculated by the spectrum similarity calculation module as the molecular structure of the target substance, and identifies and calculates the position of hydrogen substituted by deuterium in the corresponding molecular structure.

[0140] This is to find the peak of single peak and multiple peak mixture in the peak Kj of the key spectrum, find the peak corresponding to Qi of the best similar spectrum, and find the corresponding hydrogen in the final estimated molecular structure of the corresponding candidate molecular structure to identify the position of hydrogen substituted by deuterium in the molecular structure of the target substance.

[0141] (8) Hydrogen structure feature extraction module 80

[0142] This is a module for identifying and extracting structural characteristic information of hydrogen from each candidate molecular structure of the target substance calculated by the candidate molecular structure acquisition module. As mentioned above, the structural characteristic information to be identified is the number of isomorphic hydrogens, the number of hydrogens connected to the first adjacent atom, the number of hydrogens connected to the second adjacent atom, and the identification information of other hydrogens that may have COSY peaks.

[0143] Therefore, by extracting structural feature information of each hydrogen from each candidate molecular structure, it is determined whether each hydrogen of the candidate molecular structure can be assigned to each peak of the spectrum based on the feature information of the hydrogen of each candidate molecular structure in the peak assignment validity determination module.

[0144] The hydrogen structure feature extraction module of the present invention may be composed of a data input module, which uses a known algorithm for extracting hydrogen structure information to identify hydrogen structure feature information, or receives hydrogen structure feature information identified using a known algorithm.

[0145] (9) Peak allocation validity determination module 90

[0146] The peak allocation validity determination module is a module that determines whether each hydrogen can be assigned to each peak of the spectrum based on the structural feature information of each hydrogen extracted from the candidate molecular structure according to the above formula 1. The peak allocation validity determination module is composed of software that obtains the optimal solution according to the above formula 1.

[0147] The peak assignment validity determination module determines that if the optimal solution according to Formula 1 is not obtained, the corresponding candidate molecular structure is not suitable for the given spectrum.

[0148] (10) Valid candidate molecular structure output module 100

[0149] As a result of verifying the candidate molecular structures calculated by the candidate molecular structure acquisition module in each of the hydrogen structure feature extraction module and the peak allocation validity determination module, the valid candidate structure output module excludes candidate molecular structures that are not suitable for a given spectrum and outputs a list of remaining candidate molecular structures as a candidate molecular structure list. That is, the valid candidate molecular structure output module updates the candidate molecular structure list by using the result of the peak allocation validity determination module.

[0150] Generally, in order to determine the suitability of a candidate molecular structure, each hydrogen position of each candidate molecular structure is compared with the shape of each peak of the spectrum to determine whether the candidate molecular structure is a valid structure, so as the molecular structure becomes larger and more complex, the difficulty and time required increase exponentially. However, according to the present invention, by converting hydrogen that can be distinguished by a predetermined rule into a problem that can be solved by optimizing only whether hydrogen can correspond to each peak of the NMR spectrum of the target substance, the suitability of the candidate molecular structure can be determined very quickly, and the suitability of a large aromatic compound can also be accurately and quickly determined by utilizing 1H-NMR information as well as 2D 1H-1H NMR information.

[0151] Furthermore, typically, when comparing an actual spectrum with a virtual spectrum for molecular structure estimation of an unknown substance, a method is used in which similarity is evaluated to be high when signals of similar intensity are located at similar X-axis positions. Since the positions and intensities of the peaks are not significantly different from each other when a large number of peaks are concentrated in a small area (as in the case of aromatic macromolecules) and when the candidate molecular structures are similar to the extent that only the substitution positions of the functional groups are different, the above typical methods are inaccurate in determining similarity.

[0152] Therefore, the present invention makes it possible to establish a system for accurately determining the similarity of candidate structures of aromatic polymers by comparing the spectra of candidate molecular structures using not only the x-coordinates and y-coordinates of 1H-NMR peaks but also multiplicity and 2D 1H-1H COSY NMR information.

[0153] Meanwhile, the names of reference numerals used in the present invention are as follows.

[0154] 10 spectrum acquisition device 20 spectrum information extraction module

[0155] 30Candidate molecular structure acquisition module 40Virtual spectrum calculation module

[0156] 50 Spectrum similarity calculation module 60 Deuterium substitution position calculation module

[0157] 70 Final estimated structure output module 80 Hydrogen structure feature extraction module

[0158] 90 Peak assignment validity determination module 100 Valid candidate molecular structure output module

Claims

1. A method for predicting the molecular structure of a target substance from its 1H-NMR spectrum and 1H-1H COSY spectrum, the method comprising: A key spectrum acquisition step, wherein the key spectrum acquisition step is used to acquire a 1H-NMR spectrum and a 1H-1HCOSY spectrum of the target substance as key spectra; A step for obtaining key spectrum information including peak position, peak splitting value, peak integral value and cosy information of the key spectrum; a candidate molecular structure acquisition step, wherein the candidate molecular structure acquisition step is used to acquire molecular structure information of a plurality of candidates of the target substance; A step for creating a candidate molecular structure list, wherein the candidate molecular structure list contains the acquired molecular structure information of the plurality of candidates; A candidate spectrum information acquisition step, wherein the candidate spectrum information acquisition step is used to acquire a virtual spectrum corresponding to each candidate molecular structure from a plurality of candidate molecular structures in the candidate molecular structure list as a spectrum of each candidate molecular structure, and acquire information of the virtual spectrum; A spectrum comparison step, wherein the spectrum comparison step is used to compare the key spectrum information with the candidate spectrum information to calculate the similarity between the key spectrum and the candidate spectrum; as well as The target substance molecular structure determination step is used to obtain the candidate spectrum with the highest similarity in the spectrum comparison step, and select the molecular structure of the corresponding candidate substance as the molecular structure of the target substance.

2. The method according to claim 1, further comprising: After the candidate molecular structure list creation step, A hydrogen structure characteristic information extraction step, wherein the hydrogen structure characteristic information extraction step extracts the structural characteristic information of each hydrogen from the candidate molecular structure of the target substance; a peak assignment possibility determination step, wherein the peak assignment possibility determination step verifies whether all hydrogens of each candidate molecular structure can be assigned to all peak positions of the key spectrum of the target substance based on the structural feature information of each hydrogen; and The candidate molecular structure list updating step only retains the candidate molecular structures that can assign all the above hydrogens to all the peak positions of the key spectrum, and excludes the remaining candidate molecular structures from the candidate molecular structure list.

3. The method according to claim 2, wherein the peak assignment possibility determination step obtains an optimal solution of Formula 1, and if the optimal solution is obtained, it is determined as a candidate molecular structure, and the candidate molecular structure list updating step updates the candidate list by excluding inappropriate candidate molecular structures: (Formula 1) For a matrix X of size M×N make minimize (If hydrogen j is not assigned to peak i, then X ij = 0, if hydrogen j is assigned to peak i, then X ij =1,j=1,2,...,M,i=1,2,...,N) <Constraints> ①For 1≤i≤N, s i ≥0,t i ≥0 ②For all 1≤i≤N, the peak integral value of hydrogen j is intg 氢 (j), and the integral value of peak i of the key spectrum is intg 峰 (i) ③X ij = 0, (if the multiplicity of hydrogen j is different from the multiplicity of peak i) ④X ij = 0, (if the peak integral value of hydrogen j is greater than the peak integral value of peak i) ⑤ For two peaks i_1 and i_2 with COSY peaks, there must be at least one COSY peak among the set of all hydrogens assigned to i_1, C_hydrogen(i1), and the set of all hydrogens assigned to i_2, C_hydrogen(i2).

4. The method according to any one of claims 1 to 3, wherein the key spectrum information and the candidate spectrum information include: COSY peaks, where the COSY peaks are pairs of each peak in each spectrum and another peak related to the said peak; Peak integral values, where the peak integral values are the areas of each peak; The position values (hydrogen atom displacement values) of each peak; and Peak splitting values, where the peak splitting values are the number of small peaks that make up each peak.

5. The method according to claim 4, wherein the spectrum comparison step includes a peak correspondence step of allocating peaks to satisfy the following formula 2 when the peak in the key spectrum is Kj and the peak in the candidate spectrum is Qi: (Formula 2) Integral value of Qi = <Integral value of Kj Splitting value of Qi = Splitting value of Kj.

6. The method according to claim 5, wherein the peak correspondence step corresponds to allocating one or more hydrogens having a COSY relationship with Qi to another peak having a COSY relationship with Ki.

7. The method according to claim 6, wherein the target substance molecular structure determination step includes: Final candidate molecular structure determination step, which determines the molecular structure of the target substance as the molecular structure corresponding to the candidate spectrum obtained as the most similar to the key spectrum in the candidate spectrum; And Deuterium substitution position estimation step, which identifies the hydrogen corresponding to the peak in the estimated molecular structure of the target substance where a singlet and a multiplet are mixed in the peak in the key spectrum as the hydrogen substituted by deuterium.

8. A method for determining similarity by comparing two NMR spectra, the method including: Peak information extraction step, which extracts peak information for spectrum comparison from each peak of the first spectrum as the key spectrum and the second spectrum as the candidate spectrum; Peak correspondence step, which matches each peak of the first spectrum and the second spectrum; Spectrum similarity calculation step, which calculates the similarity between the first spectrum and the second spectrum by comparing the peak information of the matched peaks of the first spectrum and the second spectrum.

9. The method according to claim 8, wherein the peak information includes: COSY peaks, where the COSY peaks are pairs of the corresponding peak and another peak related to the said peak; Peak integral values, where the peak integral values are the areas of each peak; The position values (hydrogen atom displacement values) of each peak; and Peak splitting values, where the peak splitting values are the number of small peaks that make up each peak.

10. The method according to claim 9, wherein the peak correspondence step matches each peak to satisfy the following formula 3: (Formula 3) Integral value of Qi = <Integral value of Kj Splitting value of Qi = Splitting value of Kj.

11. The method according to claim 9, wherein in the spectrum similarity calculation step, the similarity is calculated by the objective function of the following formula 4: (Formula 4) (M and N are the number of peaks of the query spectrum and the key spectrum respectively, X is an M×N binary matrix, and if the query peak Qi is not assigned to the key peak Kj, then X ij = 0, and if the query peak Qi is assigned to the key peak Kj, then X ij = 1. Displacement 查询 (i) and displacement 关键 (j) represent the chemical shifts of Qi and Ki respectively. ) (Constraint conditions) ① (intg 查询 (i) is the integral value of the query peak Qi, and intg 关键 (j) is the integral value of the key peak Kj) ②If the multiplicity of Kj is different from the multiplicity of Qi, then X ij =0 ③If the integral value of Qi is greater than the integral value of Kj, then X ij =0 ④ For two key peaks Ki_1 and Ki_2 with COSY peaks, there must be at least one COSY peak in all query peak sets C_query(Ki_1) assigned to Ki_1 and all query peak sets C_query(Ki_2) assigned to Ki_2.

12. A system for estimating a molecular structure of a target substance, the system comprising: A spectrum acquisition device, wherein the spectrum acquisition device acquires a 1H-NMR spectrum and a 1H-1H NMR spectrum of the target substance from the NMR device as key spectra of the target substance; A candidate molecular structure acquisition module, which acquires the candidate molecular structures of the target substance and calculates a candidate molecular structure list; A virtual spectrum calculation module, wherein the virtual spectrum calculation module calculates a virtual spectrum from the candidate molecular structures included in the candidate molecular structure list as a candidate spectrum; a spectrum information extraction module, the spectrum information extraction module extracting spectrum information from each of the key spectrum and the candidate spectrum; A spectrum similarity calculation module, wherein the spectrum similarity calculation module calculates the similarity between the key spectrum and the candidate spectrum by using the spectrum information of the key spectrum and the candidate spectrum; as well as A final estimated structure output module is used to output the candidate molecular structure corresponding to the candidate spectrum with the highest similarity as the final estimated molecular structure of the target substance.

13. The system of claim 12, further comprising: A hydrogen structure feature extraction module, which identifies and extracts structural feature information of each hydrogen from each candidate molecular structure of the target substance; a peak allocation validity determination module, which verifies whether each hydrogen of the candidate molecular structure can be allocated to all peaks in the spectrum based on the structural feature information of each hydrogen identified from the candidate molecular structure; and An effective candidate molecular structure output module is used to update the candidate molecular structure list by using the result of the peak assignment validity judgment module.

14. The system according to claim 12 or claim 13, wherein the spectral information extracted from the spectral information extraction module includes the following information: (a) Peak position (chemical shift value): the x-axis value of each peak in the one-dimensional spectrum (b) Peak splitting value (multiplicity (split)): the number of small peaks that make up each peak (for example, in the spectrum of Figure 1(a) , the splitting of the leftmost peak is 2, and the splitting of the rightmost peak is 3) (c) Peak integral value: the area value of each peak (d) Cosy information (COSY): information of all peak pairs where cosy peaks are observed.

15. The system according to claim 14, wherein the similarity calculation module calculates the similarity between the key spectrogram and the candidate spectrogram by using the objective function in the following formula 5: (Formula 5) (M and N are the number of peaks of the query spectrum and the key spectrum respectively, X is an M×N binary matrix, and if the query peak Qi is not assigned to the key peak Kj, then X ij = 0, and if the query peak Qi is assigned to the key peak Kj, then X ij = 1. Displacement 查询 (i) and displacement 关键 (j) represent the chemical shifts of Qi and Ki respectively. ) (Constraints) ① (intg 查询 (i) is the integral value of the query peak Qi, and intg 关键 (j) is the integral value of the key peak Kj) ②If the multiplicity of Kj is different from the multiplicity of Qi, then X ij =0 ③If the integral value of Qi is greater than the integral value of Kj, then X ij =0 ④ For two key peaks Ki_1 and Ki_2 with COSY peaks, there must be at least one COSY peak in all query peak sets C_query(Ki_1) assigned to Ki_1 and all query peak sets C_query(Ki_2) assigned to Ki_2.

16. The system according to claim 15 further includes a deuterium substitution position calculation module, which finds Qi of the query spectrum of the final candidate molecular structure assigned to Kj, where single peaks and multiple peaks are mixed in the peaks of the key spectrum, and identifies the hydrogen of the corresponding final candidate molecular structure as hydrogen substituted by deuterium.

Images