Method and system for obtaining molecular structure of parent molecule from fragment ions from LC-MS / MS

Abstract

Disclosed herein are methods and systems that acquire mass spectrum peak data for fragment ions from a mass spectrum of fragment ions generated in an LC-MS / MS measurement instrument, acquire molecular structure and fragment ion map data for fragment ions by using the acquired data, and determine fragment ion map data for fragment ions. And a candidate molecular structure of the target substance is obtained by assembling the fragment ion diagram.

Description

Technical Field

[0001] The present invention relates to a technique for determining and analyzing the molecular structure of a parent molecule (original compound) based on information on fragment ions generated during liquid chromatography-tandem mass spectrometry (LC-MS / MS) analysis. Background Art

[0002] When analyzing the structure of a material by liquid chromatography-tandem mass spectrometry (LC-MS / MS) analysis, strong energy is applied to the sample molecules to ionize and fragment the molecules, and then the mass is measured based on the velocity of each fragment ion. In this case, the structure of the parent molecule can be estimated based on the elemental composition and detection intensity of each fragment ion and the elemental composition of the parent molecule.

[0003] However, in order to achieve the estimation, since the structure of the parent molecule must be estimated based on the elemental composition and detection intensity of each fragment ion and the elemental composition of the parent molecule, it is necessary to be able to estimate the structure of the individual fragment ions and then draw the structure of the parent molecule containing the structures of all fragment ions as partial structures.

[0004] However, it is usually impossible to accurately estimate the structure of the fragment ions using only the information provided by the MS / MS spectrum, and each fragment ion has multiple possible candidate structures. Therefore, in order to use the structure of the fragment ions to estimate the structure of the parent molecule, it is necessary to explore both the number of structural combinations of the individual fragment ions and their assembly situations, which leads to the problem that the computation time grows exponentially relative to the average number of nodes in the graph to be assembled. For example, the number of subgraphs in a graph of size N is 2 N , and since subgraph isomorphism is also an NP-complete problem, it is impossible to solve the combinatorial problem for all subgraphs in polynomial time.

[0005] The prior art related to the present invention is as follows:

[0006] JP 2019-100891 A (June 24, 2019)

[0007] High-Resolution Liquid Chromatography Tandem Mass SpectrometryEnables Large Scale Molecular Characterization of Dissolved Organic Matter, Danial Petras et al., METHODS, December 12, 2017 doi:10.3389 / fmars.2017.00405 Summary of the invention

[0008] Technical issues

[0009] The present invention aims to provide a method and system for estimating the molecular structure of a target substance using liquid chromatography-tandem mass spectrometry (LC-MS / MS).

[0010] In order to solve the above problems in the prior art, the present invention aims to provide an algorithm capable of solving the problem of combining fragment ions calculated for a target substance in polynomial time and a system applying the algorithm, thereby making it possible to estimate the molecular structure of the target substance.

[0011] Technical Solution

[0012] In order to solve the above problems, according to the present invention, a method for obtaining the molecular structure of a target substance is provided, the method comprising a fragment ion candidate structure deriving step: deriving the candidate structure of the fragment ion by inputting the mass spectrum of the fragment ion obtained from the LC-MS / MS analysis of the target substance into a predetermined spectrum-molecular structure database; a fragment ion graphics step: converting the obtained candidate structure of the fragment ion into fragment ion map data; a fragment ion map assembling step: assembling the fragment ion candidate structure by using the fragment ion map data to obtain the candidate molecular structure of the target substance; and a target substance molecular structure determining step: determining the target substance molecular structure from the candidate molecular structure of the target substance in the fragment ion map assembling step.

[0013] To this end, the present invention applies a method for representing the candidate structures of fragment ions as a fragment ion graph by representing the invariant atomic groups contained in the fragment ions and the connecting atoms connected to the invariant atomic groups as nodes, and representing information about the atomic groups or connecting atoms represented by each node as attribute information about each node.

[0014] When assembling fragment ion maps, the present invention provides such a method: a tree node assignment step (S31): assigning the fragment ion maps corresponding to each peak of the mass spectrum to a tree node variable for a tree search operation; an access node selection step (S32): sequentially selecting and accessing tree nodes of a first depth, excluding tree nodes set as stop nodes; a subtree node creation step (S33): setting the accessed tree node as a parent node, and connecting all tree nodes corresponding to subsequent peaks as child nodes; an assembly intermediate creation step (S34): creating an assembly intermediate in a child node by sequentially assembling the fragment ion map assigned to the child node with the assembly intermediate map of the intermediate node variable assigned to the parent node, assigning the assembly intermediate to the assembly intermediate node variable corresponding to the corresponding child node, and setting the corresponding child node as a stop node when a predetermined assembly intermediate binding condition is not met; and obtaining the final assembly intermediate created by repeating the access node selection step to the assembly intermediate setting step until there are no more subtree nodes to be created as a candidate molecular structure of the target substance.

[0015] In addition, as a method for creating assembly intermediates by assembling fragment ion maps, the present invention provides such a map assembly method, which includes: an assembly target fragment ion selection step: selecting the assembly intermediate map of the parent node and the fragment ion map of the child node as the target fragment ion map data to be assembled and the input assembly fragment ion map to be assembled, respectively; a sub-map data extraction step: extracting all sub-maps from the input fragment ion map data; and a sub-map combination step: sequentially combining all sub-maps into a target fragment ion map.

[0016] In addition, according to the present invention, a system for obtaining the molecular structure of a target substance is provided, the system comprising: an LC-MS / MS measurement unit, the LC-MS / MS measurement unit being configured to separate the target substance into fragment ions and generate a mass spectrum of the fragment ions; a spectrum preprocessing module, the spectrum preprocessing module being configured to obtain mass spectrum peak data of the fragment ions from the mass spectrum of the fragment ions, obtain the molecular structure of the fragment ions from a spectrum-molecular structure database using the mass spectrum peak data, and perform fragment ion graphing on the molecular structure of the fragment ions to generate fragment ion map data; a map assembly module, the map assembly module being configured to derive a candidate molecular structure of the target substance by receiving the fragment ion map data and assembling the fragment ion maps; and a spectrum-molecular structure database, the spectrum-molecular structure database being configured to output the molecular structure of the fragment ions by receiving mass spectrum peak data on the fragment ions.

[0017] Beneficial Effects

[0018] According to the present invention, the molecular structure of a target substance can be estimated by applying a method of deriving fragment ion maps corresponding to respective peaks of a fragment ion mass spectrum and assembling the fragment ion maps via a tree search technique. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The following drawings attached to the specification illustrate preferred embodiments of the present invention by way of example and are used to enable further understanding of the technical aspects of the present invention together with the detailed description of the present invention given above, and therefore the present invention should not be interpreted solely by the contents in such drawings.

[0020] Figure 1 is a block diagram of a system for acquiring the molecular structure of a parent molecule from fragment ions according to the present invention.

[0021] Figure 2 is an example of a mass spectrum obtained in the present invention.

[0022] Figure 3 is a flow chart showing the procedure of the present invention for obtaining the molecular structure of a parent molecule from fragment ions.

[0023] Figure 4 is a graphical diagram for describing fragment ions according to the present invention.

[0024] Figure 5 is a diagram for describing a process of assembling fragment ion maps corresponding to mass spectrum peaks according to the present invention.

[0025] Figure 6 is a flow chart illustrating a process of assembling fragment ion maps corresponding to mass spectral peaks according to the present invention.

[0026] Figure 7 is a diagram showing a process of combining fragment ion maps according to a search tree procedure according to the present invention.

[0027] Figure 8 is a diagram showing an example of assembling subgraphs into an assembly target graph according to the present invention.

[0028] Fig. 9 is a flow chart showing a procedure for assembling subgraphs into an assembly target graph according to the present invention.

[0029] Fig.10 is a diagram showing the process of assembling by applying subgraph isomorphism.

[0030] Fig.11 is an example of a pseudo code for implementing a subgraph assembly algorithm according to the present invention.

[0031] Fig.12 is an example of deriving the structure of a parent molecule from the graph assembly result according to the present invention. DETAILED DESCRIPTION

[0032] For such parent molecule structure estimation, it is necessary to estimate the structure of the individual fragment ions and then draw the parent molecule structure containing all of them as partial structures. However, since it is impossible to accurately estimate the structure of the fragment ions using only the information provided by the MS / MS spectrum, each fragment ion corresponds to a variety of possible candidate structures, so in order to use the candidate structures to estimate the structure of the parent molecule, it is necessary to explore both the number of structural combinations of the individual candidate fragment ions and the number of situations in which they are assembled.

[0033] To this end, the invention allows to superimpose the fragment structures on each other and assemble them, but estimates the structure of the parent molecule by an algorithm that extracts the number of all cases where the molecules assembled from all fragments have the same elemental composition as the parent molecule.

[0034] In the following, reference is made to Figures 1 to 12 , a method and system for acquiring the molecular structure of a target substance (parent molecule) from fragment ions generated during liquid chromatography-tandem mass spectrometry (LC-MS / MS LC-MS / MS) analysis of the target substance according to the present invention will be described.

[0035] 1. Parent molecule molecular structure acquisition system

[0036] like Figure 1 As shown, the system of the present invention includes an LC-MS / MS measurement unit 100 , a spectrum preprocessing module 200 , a graph assembly module 300 , and a spectrum-molecular structure database (DB) 400 .

[0037] The LC-MS / MS measurement unit 100 includes a typical LC-MS / MS measurement instrument (liquid chromatography-tandem mass spectrometry (LC-MS / MS)). This device separates a target substance (parent molecule) into fragment ions and generates a mass spectrum of the fragment ions.

[0038] Figure 2 The example of the mass spectrum generated by fragment ions is shown. The mass spectrum of fragment ions can be obtained as a figure consisting of a mass axis (molecular weight) and an ionic strength axis representing the mass of each fragment ion with a mass-to-charge ratio (m / z), and when the specific mass (m / z) in the mass spectrum of fragment ions is found in the figure, the molecular weight of fragment ions can be calculated. In this case, since the charge of fragment ions is generally 1, it can be said that the mass and molecular weight of fragment ions are almost the same, and therefore the m / z value of fragment ions can be used as molecular weight. For example, when the m / z value of fragment ions is 200, it can be said that the molecular weight of fragment ions is 200 Daltons (g / mol). In fragment ion spectrum information, molecular weight information is used to obtain the molecular structure of fragment ions from spectrum-molecular structure DB 400.

[0039] The ion intensity value is related to the amount of fragment ions generated. A high ion intensity value means that the fragment is cut many times, and generally, since the fragment ions located in the center of the compound molecule have high stability and are less likely to be cut into fragment ions, when deriving the final parent molecule candidate structure, the fragment ions with high ion intensity values ​​are located in the center, and therefore the structure that must break multiple bonds is likely to be a candidate structure that does not conform to the actual parent molecule structure.

[0040] The spectrum preprocessing module 200 obtains the mass spectrum peak data (molecular weight, ion intensity value of each fragment ion) of each fragment ion constituting the target substance from the LC-MS / MS measurement unit 100, obtains the molecular structure of the fragment ion by inputting the data into the spectrum-molecular structure DB, and reconstructs the molecular structure of the obtained fragment ion into fragment ion map data. The spectrum-molecular structure DB is a database containing known mass spectrum data and corresponding molecular structure data about molecules, and known spectrum-molecular structure DBs include NIST Chemistry WebBook, NIST Mass Spectral Library, METLIN, PubChem, ChemSpider, etc., but the scope of the present invention is not limited to the use of a specific database, and as a database that carries its own data, any database containing mass spectrum data and corresponding molecular structure data about a given molecule can be applied to the system of the present invention.

[0041] The spectrum preprocessing module 200 receives molecular structure data on fragment ions corresponding to mass spectrum data on fragment ions input from the spectrum-molecular structure DB, and reconstructs the received data into mapping data.

[0042] In the present invention, "graph" refers to a specific molecular structure represented in the form of a graph, and "graph data" refers to the conversion of a graph into numerical data that can be processed by computer operations. Graph data may include nodes constituting the graph, edges connecting the nodes, node characteristics, and edge characteristic data, and may be represented as matrix data, such as an adjacency matrix representing the same, etc., but is limited to a specific data format.

[0043] According to the fragment ion graph of the present invention (which reconstructs the molecular structure of the fragment ions into graph data), each invariant atomic group is represented as a node, atoms that are not invariant atomic groups are represented as other nodes connected to the invariant atomic group nodes, and there are edges connecting each node (node: this is a node representing a molecular structure, and is different from the "tree node", "parent node" and "child node" described below). The fragment ion graph data can be graph data including variables and node attribute data of each node representing the fragment ion graph, and edge variables and edge attribute data connecting the nodes.

[0044] The graph assembly module 300 can receive the fragment ion graph data obtained by graphing the fragment ions, assemble the fragment ion graph according to the fragment ion assembly method of the present invention described below, and calculate the assembled intermediate graph data (which is the fragment ion combined molecule graph data), and also output the assembled intermediate graph data as the fragment ion combined molecular structure, that is, the candidate structure of the estimated parent molecule. The graph assembly module 300 may include a parent molecule structure determination module (not shown) to determine the final parent molecule structure from the candidate structures through the possibility evaluation operation described below.

[0045] In the present invention, the spectrum preprocessing module 200 and the graph assembly module 300 can each be a separate data operation module, or can be a single computing device or computer device. In this case, each computing module, computing device or computer device can be configured to include a storage device and a computing device for storing, reading and calculating a computer-implemented algorithm for performing the graphical and graph assembly method of the fragment ions described below.

[0046] 2. Methods for obtaining the molecular structure of the parent molecule

[0047] Reference Figure 3 , a method for obtaining the molecular structure of a parent molecule according to the present invention will be generally described.

[0048] (1) Fragment ion candidate structure acquisition step (S100)

[0049] First, candidate structures of fragment ions were obtained by performing the following procedure.

[0050] In order to obtain the molecular structure of a given target substance (target compound, parent molecule) having an unknown molecular structure, firstly, LC-MS / MS analysis is performed on a sample containing the target substance.

[0051] In LC-MS / MS analysis, the target substance in the sample (i.e., the target compound in the sample) is separated by liquid chromatography (LC), the target compound is input into the MS / MS device, fragment ions are generated, and a mass spectrum (MS / MS spectrum) of the generated fragment ions is generated, wherein the present invention begins the process of estimating the molecular structure of the parent molecule by acquiring the mass spectrum data about the fragment ions generated at this time.

[0052] Next, the mass spectrum of the obtained fragment ions is input into the spectrum-molecular structure DB 400 to obtain candidate structures of the fragment ions matching the corresponding spectrum. Figure 4As shown in the example in , since it is difficult to know in advance which fragment ions will correspond to each peak in the mass spectrum, a list of all fragment ions of the same elemental composition is retrieved and assembled for each of them. The multiple fragment ions that constitute the list of fragment ions corresponding to each peak in the mass spectrum are called candidate fragments.

[0053] (2) Fragment ion patterning step (S200)

[0054] Next, this is a step of obtaining fragment ion map data by graphing the molecular structure of the candidate fragment obtained from the database and converting the fragment ion map into data that can be processed by computer operation. As described below, converting the fragment ion map into fragment ion map data that can be processed by computer operation is to represent the fragment ion map as a graph adjacency matrix including graph nodes and their attribute information, edges and edge attribute information, a node attribute matrix and an edge attribute matrix, which is well known to those skilled in the art.

[0055] In the present invention, the fragment ion candidate structure is referred to as a candidate fragment, and when it is represented as graph data, the invariant atomic group (which is always detected in a combined form on the MS / MS spectrum) contained in the fragment ion is represented as a single node. An atom that does not belong to the invariant atomic group is represented as a node. Information about each invariant atomic group and atom is represented as attribute information about each node.

[0056] For example, Figure 4 As shown, benzene is an invariant atomic group, which is always detected in its complete form and regarded as one node. However, for carbazole, the central pentagonal structure is sometimes decomposed and detected as a combination of benzene and nitrogen atoms, and therefore, carbazole is not regarded as an invariant atomic group, but is represented as a connection between a benzene node and a nitrogen node, which are invariant atomic groups.

[0057] (3) Fragment ion assembly step (S300)

[0058] Next, the fragment ion maps representing candidate structures of the fragment ions are assembled to obtain candidate structures of the parent molecule.

[0059] However, it is difficult to know in advance which fragment ions will correspond to the peaks of the MS / MS spectrum. Therefore, all lists of fragment ions with elemental compositions corresponding to each peak of the spectrum are retrieved from the spectrum-molecular structure DB 400, and then assembled for each of them. 1 , p 2 , ..., p N}, and there exists a i The corresponding n iWhen there are candidate fragment ions, the total number of fragment ion combinations that must be assembled becomes And the calculation takes a long time.

[0060] Therefore, in the present invention, when any two combinations share most of the fragment ion combinations, it is determined that it would be inefficient to assemble each of them from scratch, and a method is applied to minimize repeated path visits by conducting the assembly process in the form of a search tree. During the tree search process, pruning and excluding nodes that may no longer be assembled eliminates the need to visit all nodes branching from the node, thereby greatly reducing the time. This will refer to Figure 5 Give a description.

[0061] In the present invention, N peaks are listed in order of size, and then a search tree is created, for which a search node is the behavior of selecting one of the fragment ion candidates. A node with depth i corresponds to the behavior of selecting which fragment ion to assemble at the i-th peak. Applying this principle, the present invention performs a tree search in the following order. In this way, the tree search process assigns the fragment ion map corresponding to each peak to the corresponding tree node variable and performs a tree search operation.

[0062] Figure 5 is a diagram showing the process of selecting a combined fragment ion when there are 5 peaks in the mass spectrum and the numbers of candidate fragment ions corresponding to the peaks are 3, 4, 2, 1, and 1, respectively. Figure 5 and Figure 6 , the specific procedure of the tree search operation will be described.

[0063] exist Figure 5 In the example, three tree nodes are assigned to peaks 1 , and assigning four tree nodes to peaks 2 Each tree node variable corresponds to the fragment ion map data of each fragment ion. Figure 5 In the example, the number of possible combinations of the fragment graph is 3*4*2*1*1=24 cases, and according to the combination method of the present invention, the number of possible combinations is reduced, thereby saving computing resources.

[0064] (a) Tree Node Assignment Step (S31)

[0065] To perform a tree search operation, the fragment ion maps corresponding to each peak of the mass spectrum are assigned to each tree node variable of the search tree. The root node of the search tree is set to have an empty map. The corresponding fragment ion map is assigned to each node of each peak, and each fragment ion map can be represented as g_frag i (k) (k represents the peak number, and i represents the number of the fragment pattern of the peak).

[0066] The fragment ion map corresponding to each peak is assigned to the tree node of each depth of the search tree. For example, the starting node of the search tree is assigned to the root node with an empty map, and the fragment ion map corresponding to peak_b is assigned to the tree node at depth b.

[0067] like Figure 7 As shown, in the search tree, the starting node can be TN 根 , and other tree nodes can be represented as TN a (b) variables (b is the depth of the search tree, and a is the order of the tree nodes at that depth).

[0068] (b) Access Node Selection Step (S32)

[0069] This is the step of sequentially selecting and accessing the tree node to be accessed in the tree node. In fact, this step can be realized with a computer algorithm in which a predetermined tree node selection pointer sequentially points to each of the tree nodes.

[0070] The fragment graph assembly procedure described below is performed by selecting the first tree node of the first peak, the graph assembly procedure described below is performed by sequentially selecting all corresponding tree nodes of one peak, and then the tree node of the next peak is selected. Meanwhile, the selection order of the tree nodes is not limited thereto, and it is sufficient to go through the process of selecting all tree nodes of all peaks regardless of their order.

[0071] The selection of tree nodes is performed according to the procedure described below, excluding the tree nodes marked as "abort" according to a predetermined condition. The indication of "abort" can be displayed as "abort" for the tree node by changing the value of a specific flag in the tree node variable, and the indication of "abort" does not mean to stop the procedure, but means not to assemble the fragment graph corresponding to the tree node.

[0072] The access node selection step may start from accessing the first parent node among the top parent nodes. In the subsequent iteration step, when the access node is marked as "aborted", the subsequent procedure is not performed, but the next node is accessed.

[0073] (c) Child node creation step (S33)

[0074] For visited tree nodes, child nodes are created corresponding to the respective fragment ion maps of the peaks following the peak to which the respective tree node corresponds. This is the step of generating the number of cases in which the respective fragment ion candidates that can correspond to the subsequent peaks are assembled in the maps assembled so far.

[0075] (d) Assembly intermediate graph generation step (S34)

[0076] This is a step of generating an assembly intermediate graph at each child node by assembling the corresponding fragment ion graph (child node graph) with the assembly intermediate graph of the parent node at each child node. Each tree node can be assigned a corresponding assembly intermediate node variable, and as a child node, can store the assembly intermediate graph of its parent node and the assembled assembly intermediate graph accordingly.

[0077] In this case, in the assembly intermediate graph generation step (S34), at each child node, the assembly intermediate graph stored in the assembly intermediate node variable of the parent node and the child node graph are assembled, and the graph data as the assembly result is stored in the assembly intermediate node variable of the child node for assembly with the child node graph of the next depth. After the assembly intermediate graph is generated for all child nodes, when the corresponding peak is not the last peak, each child node is visited, and the process returns to the visit node selection step (S32).

[0078] Meanwhile, among the child nodes satisfying predetermined conditions to be described below, the child nodes that cannot create any assembly intermediate with the parent node are added to the list of aborted nodes.

[0079] (e) Step of obtaining a candidate structure of a parent molecule (S35)

[0080] This is a step of obtaining a final updated assembly intermediate graph by repeating steps S32 to S34 until a node of depth N corresponding to the last peak of the peak list is created, and obtaining a candidate molecular structure of a parent molecule (target substance) therefrom.

[0081] for Figure 5 For examples, refer to Figure 7 The above-mentioned processes S31 to S35 are described as examples. Figure 5 In the example, due to the existence of p 1 to p 5 , and the fragment ions corresponding to each peak are 1, 1, 2, 4 and 3, respectively, so the number of graph combinations to be considered in the conventional method is 1*1*2*4*3=24 types. However, according to the stop node deletion method used in the present invention, the number of combinations is reduced, thereby improving the calculation efficiency.

[0082] The order of assembly of the peaks can be chosen arbitrarily, and Figure 7 In the example, the case where the sequence p5->p4->p3->p2->p1 is assembled will be described. a b is assigned to each tree node variable in the search tree, where a represents the node sequence number and b represents the tree depth. For example, TN 2 (1) refers to the second node at depth 1, and TN12 (2) refers to the 12th node at depth 2. Similarly, with each tree node TN a (b) The corresponding assembly intermediate graph is denoted as G a (b) .

[0083] Figure 7 (a) in FIG. 1 shows the initial start of the tree search, that is, the step of selecting the root node as the access node ( S32 ).

[0084] In the next step, at depth 1, the peaks are assigned 5 The corresponding fragment ion map frag 1至3 (5) The child node (S33) of is used as the child node of the access node. Figure 5 In the example of 5 , thus creating three child nodes. In the next step, each child node is assembled with the root node to generate an assembly intermediate graph (S34). Since the root node has no corresponding peak or fragment ion, the assembly intermediate graph generated by the three child nodes is aligned with the peak at depth 1. 5 That is, at depth 1, the tree node TN 1至3 (1) is created as a child of the root node, and the fragment ion graph frag 1至3 (5) They are assigned to the tree nodes TN 1至3 (1) The corresponding assembly intermediate G 1至3 (1) (The fragment ion map is represented as frag a (b) , where b is the peak number corresponding to the fragment ion, and a is the fragment ion map number of the corresponding peak).

[0085] because Figure 7 The peak in the procedure shown in (a) is not the final peak, so Figure 7 As shown in (b) of FIG. 1 , by returning to the access node selection step, the peak corresponding to the generated 5 The corresponding tree node is used as the access node (S32), and for the peak 5 Each tree node is created to be assigned with the peak 4 The corresponding fragment ion graph child node (S33). Since the four fragment ions correspond to the peak 4 , so the four child nodes are assigned to the peak 5Now, at each of the child nodes (12 in total), the assembly intermediate graph of the parent node and the fragment ion graph corresponding to the child node are combined to generate an assembly intermediate graph corresponding to the child node in this step (S34). 1 (2) The corresponding assembly intermediate G 1 (2) frag 1 (5) +frag 1 (4) , and with TN 2 (2) The corresponding assembly intermediate G 2 (2) frag 1 (5) +frag 2 (4) .

[0086] When the molecular weight in the assembled intermediate graph generated by such a combination becomes larger than the molecular weight of the parent molecule before the target substance is input into the mass spectrometer, the combination does not seem to satisfy the binding condition, and thus the node is set as a stop node. Figure 7 , shows a situation where due to TN 3 (2) TN 4 (2) TN 6 (2) TN 8 (2) TN 9 (2) and TN 11 (2) The assembly intermediate graph assembled in TN does not satisfy the predetermined combination described below, so TN 3 (2) TN 4 (2) TN 6 (2) TN 8 (2) TN 9 (2) and TN 11 (2) Set to abort node.

[0087] Since the last peak has not been reached, the process returns to the access node selection step (S32), as shown in FIG. Figure 7 As shown in (c) of 2 Each tree node will be 3The tree node of is created as a child node (S34), and the peak 2 In this case, even if the TN that has been previously set as the abort node is accessed 3 (2) TN 4 (2) TN 6 (2) TN 8 (2) TN 9 (2) and TN 11 (2) When , the child node creation step and the assembly intermediate creation step are not performed, so that the amount of calculation can be reduced. In this way, for all nodes TNa at depth 3 (3) , the visiting process, the child node creation process and the assembly intermediate creation process are performed, and then the visiting node selection step is performed again at the next depth. This process is repeated until the final peak is reached.

[0088] (f) Parent molecular structure determination step (S36)

[0089] Next, the best candidate molecular structure is selected from the candidate molecular structures (which are the final assembly intermediates assembled when reaching the final peak) and determined as the final parent molecular structure. The final parent molecular structure can be determined by the possibility evaluation of the candidate molecular structures.

[0090] When the candidate molecular structure is assumed to be ζ ​​and the mass spectrum of the fragment ion used to obtain the candidate molecular structure is X=x 0 ,x 1 ,...,x i (x k is a peak group consisting of continuous deprotonated peaks, that is, a set of continuous peaks with a difference of one hydrogen), in this case, the probability of the peak group X appearing due to the candidate molecular structure ζ can be expressed as the product of the occurrence probabilities of each peak in the set X due to the candidate molecular structure ζ, as shown in the following formula 1, and the possibility evaluation is to find the structure that maximizes the possibility represented by formula 1.

[0091] (Formula 1)

[0092] p(ξ|X)~Π k p(ξ|x k )

[0093] When assuming that for the peak group x k With m / zm i (k) and the normalized intensity r i(k) Each peak x i (k) =(m i (k) ,r i (k) ), the corresponding structure in the partial structure of ζ is When , the probability of the peak group can be re-expressed as the following formula 2.

[0094] (Formula 2)

[0095]

[0096] (m i (k) and r i (k) They are peak groups x k m / z and normalized intensity)

[0097] Here, in order to generate Must be fragmented i (k) times, and must be deprotonated i (k) times, and assuming that each probability is expressed as a constant independent of the structure, p f (the probability of breaking a certain bond) and p f (the probability of removing an additional hydrogen from the substructure), making the probability of each peak Maximization is equivalent to minimizing the cross entropy in Equation 3 below.

[0098] (Formula 3)

[0099]

[0100] (logp f and logp d are each a hyperparameter and are values ​​selected from the above spectrum-molecular structure database, a is the number of bonds that must be destroyed to generate the fragment ion map S in the final result map ζ, and b is the number of hydrogens additionally removed from the fragment ions)

[0101] Combining equations 1 to 3, the fitness score of each candidate structure can be defined as equation 4 below.

[0102] (Formula 4)

[0103]

[0104] (logp j and logp dare each a hyperparameter and are values ​​selected from the above spectrum-molecular structure database, a is the number of bonds that must be destroyed to generate the fragment ion map S in the final result map ζ, and b is the number of hydrogens additionally removed from the fragment ions)

[0105] In this way, the fitness of each candidate structure is calculated by calculation according to the various formulas, and based on the calculated fitness, the final candidate structure of the target substance is calculated from the final assembly intermediate map obtained by assembling the fragment ion maps.

[0106] (4) Detailed description of the assembly intermediate graph generation step (S34)

[0107] In the assembly intermediate graph generation step, the reference Figures 8 to 11 The algorithm used to generate the assembly intermediate graph is described in more detail.

[0108] Since each peak in the MS / MS spectrum represents a fragment ion obtained by decomposing the parent molecule using different methods, there is overlap between the molecular structures of the detected candidate fragments. Therefore, when combining fragment ion maps, overlap must be allowed, and the number of all possible cases of assembling a given map based on the elemental composition of the parent molecule must be extracted and assembled.

[0109] Figure 8 An example is shown in which, when assembling is performed in the order of ①→②→③→④, only two cases are picked up in each step, schematically showing a form in which assembly can be performed by allowing overlap.

[0110] Fig. 9 A specific algorithm for the assembly intermediate generation procedure (S34) is shown.

[0111] (a) Assembly target graph selection step (G10)

[0112] Two maps to be assembled are selected from the fragment ion maps representing candidate structures of the fragment ions. The selection of maps to be assembled is repeated until all fragment ion maps are selected as assembly targets and the assembly is completed.

[0113] This step is a step in which, after creating a subtree node at the access node in the previous S33, the assembly intermediate map of the parent node and the fragment ion map of the child node are respectively selected as the target fragment ion map data to be assembled and the input assembly fragment ion map to be assembled.

[0114] (b) Subgraph extraction step (G20)

[0115] The step of selecting the first fragment ion map to be assembled and assembling is referred to as step t, and the step of selecting the second map to be assembled and assembling is referred to as step t+1.

[0116] In the map to be assembled, the fragment ion map to be assembled is called the target map (current map), and the target map at step t is called g c (t) (the current graph at step t). The graph to be assembled is called the input graph (incoming graph) and is denoted as g i (Incoming graph). The set of all resulting graphs in which two graphs are assembled is called G (t+1) .

[0117] Fig.10 shows that G1 is selected as the target graph g c And select G2 as the input graph g i G1 includes node n i (i=0 to 5), and G2 includes node n k (k=0 to 3).

[0118] This is to extract the graph to be assembled (i.e., the input graph (incoming graph) g i ) of all subgraphs Si. Fig.10 It is shown that three subgraphs are extracted, with nodes (0, 1), (0, 1, 2) and (2, 3) of G2 as nodes. The algorithm for extracting subgraphs from a graph is a known algorithm.

[0119] (c) Subgraph assembly step (G30)

[0120] Next, the subgraph assembly step is performed, where the input graph g is i All subgraphs Si of are assembled sequentially into g c .

[0121] ① Partial graph initialization step (G31)

[0122] The assembly of the subgraph begins by setting the target graph (the current graph) to g c (t) , set the incoming graph to be assembled to g i , and the generated graph set G (t+1) Set to the empty set.

[0123] ②Node mapping step (G32)

[0124] Next, for the incoming graph g i Each subgraph Si(∈g i ), and then compare with the target graph g c (t) Isomorphic graph identification (subgraph isomorphism) and node mapping.

[0125] ③Node / edge combination step (G33)

[0126] When assuming that all nodes have mappings M, Si and g c (t) The subgraph isomorphism graph of k ={(v i , v C )|v i ∈S i , v c ∈g c (t)}, and the complement of subgraph Si is Si c When, for each and every node mapping M k (∈M), perform the following procedure: G c (t) Substitute g c (t+1) And Si c Add all nodes and edges of g to c (t+1) procedures; G i Does not belong to Si and Si c Add all the edges of g to c (t+1) procedures; and G c (t+1) Add to G (t+1) In this case, g c (t) Molecular weight and Si c The sum of the molecular weights of the parent molecule must be equal to or less than the molecular weight of the parent molecule. c (t) Molecular weight and Si c When the sum of the molecular weights of the two molecules exceeds the molecular weight of the parent molecule, it is determined that they cannot bind.

[0127] For g i Perform the above procedures ② and ③ for all subgraphs Si.

[0128] Fig.11 shows an example of pseudo code for performing the above procedure, and Fig.12 An example of a graph produced by combining the procedures is shown.

[0129] The names of the symbols in the drawings used in the present invention are as follows.

[0130] 100LC-MS / MS measurement unit 200spectral preprocessing module

[0131] 300 image assembly module 400 spectra-molecular structure DB

Claims

1. A method for obtaining a molecular structure of a target substance, the method comprising: The fragment ion candidate structure deriving step is: deriving the candidate structure of the fragment ion by inputting the mass spectrum of the fragment ion obtained from the LC-MS / MS analysis of the target substance into a predetermined spectrum-molecular structure database; Fragment ion mapping step: converting the candidate structures of the obtained fragment ions into fragment ion map data; Fragment ion map assembling step: assembling candidate fragment ion structures by using the fragment ion map data to obtain a candidate molecular structure of the target substance; and The target substance molecular structure determination step is to determine the target substance molecular structure based on the candidate molecular structures of the target substance in the fragment ion map assembly step.

2. The method according to claim 1, wherein in the fragment ion patterning step, the invariant atomic groups contained in the fragment ions and the connecting atoms connected to the invariant atomic groups are represented as nodes, and Information about the atomic group or the connected atoms represented by each of the nodes is represented as attribute information about each node.

3. The method according to claim 2, wherein the fragment ion map assembling step comprises: Tree node assignment step (S31): assigning the fragment ion map corresponding to each peak of the mass spectrum to the tree node variable TN a (b) (b is the depth of the search tree, and a is the order of the tree nodes at that depth) to perform a tree search operation; Visit node selection step (S32): sequentially selecting and visiting tree nodes of the first depth, excluding tree nodes set as stop nodes; Subtree node creation step (S33): the visited tree node TN a (b) Set as the parent node and all tree nodes TN corresponding to subsequent peaks a (b+1) Connect as child node; Assembly intermediate creation step (S34): by assigning a (b+1) The fragment ion map is assigned to the parent node TN a (b) sequentially assembling the assembly intermediate graphs of the intermediate node variables to create assembly intermediates in the child nodes, assigning the assembly intermediates to the assembly intermediate node variables corresponding to the corresponding child nodes, and setting the corresponding child nodes to the abort nodes when a predetermined assembly intermediate combination condition is not satisfied; as well as A final assembly intermediate created by repeating the access node selection step to the assembly intermediate setting step until there are no more subtree nodes to be created is obtained as a candidate molecular structure of the target substance.

4. The method according to claim 3, wherein the assembly intermediate creation step comprises: Assembly target fragment ion selection step: selecting the assembly intermediate map of the parent node and the fragment ion map of the child node as target fragment ion map data to be assembled and input assembly fragment ion map to be assembled, respectively; Sub-image data extraction step: extracting all sub-images from the input fragment ion image data; as well as Sub-image combination step: sequentially combining all the sub-images into the target fragment ion image.

5. The method according to claim 4, wherein in the assembly intermediate creation step, when the subgraphs are combined, when the molecular weight of the molecule corresponding to the combined graph exceeds the molecular weight of the parent molecule, it is determined that the assembly intermediate binding condition is not met, and the corresponding child node is set as the termination node.

6. The method according to claim 4, wherein in the step of determining the molecular structure of the target substance, the fitness score of the candidate molecular structure of the target substance is calculated, and the molecular structure of the target substance is determined based on the calculated fitness score.

7. A system for obtaining a molecular structure of a target substance, the system comprising: an LC-MS / MS measurement unit configured to separate the target substance into fragment ions and generate a mass spectrum of the fragment ions; a spectrum preprocessing module, the spectrum preprocessing module being configured to obtain mass spectrum peak data of the fragment ions from the mass spectrum of the fragment ions, obtain the molecular structure of the fragment ions from a spectrum-molecular structure database using the mass spectrum peak data, and perform fragment ion graphics on the molecular structure of the fragment ions to generate fragment ion map data; A map assembly module configured to derive a candidate molecular structure of the target substance by receiving the fragment ion map data and assembling the fragment ion maps; and A spectrum-molecular structure database is configured to output the molecular structure of the fragment ion by receiving the mass spectrum peak data about the fragment ion.

8. A system according to claim 7, wherein the spectrum preprocessing module performs fragment ion graphics on the fragment ion molecular structure by representing the invariant atomic groups contained in the fragment ions and the connecting atoms connected to the invariant atomic groups as nodes, and representing information about the atomic groups or the connecting atoms represented by each node as attribute information about each node.

9. A system according to claim 8, wherein the graph assembly module derives the candidate molecular structure of the target substance by assembling the fragment ion graphs corresponding to the respective peaks of the fragment ion spectra, and when the molecular weight of the molecule corresponding to the combined graph is not less than the molecular weight of the precursor ion, it is determined that the assembly intermediate binding condition is not met, and the subsequent assembly of the corresponding fragment graph is no longer performed.

10. A device for calculating the molecular structure of a target substance, the device comprising: a spectrum preprocessing module configured to receive a fragment ion spectrum of the target substance from an LC-MS / MS measuring instrument and calculate fragment ion map data by graphing the molecular structure of the fragment ions; as well as A map assembly module is configured to derive a candidate molecular structure of the target substance by receiving the fragment ion map data and assembling the fragment ion maps.

11. The device according to claim 10, wherein the spectrum preprocessing module performs fragment ion graphics on the fragment ion molecular structure by representing the invariant atomic groups contained in the fragment ions and the connecting atoms connected to the invariant atomic groups as nodes, and representing the information about the atomic groups or the connecting atoms represented by each node as attribute information about each node.

Images