Method and apparatus for regulating biological metabolic pathways

Abstract

The application provides a biological metabolic pathway regulation method and device. The method comprises the following steps: solving a wild type flux range and an overexpression type flux range of a metabolic network model; determining a reaction occurrence direction according to the overexpression type flux range; determining an operation to be performed and a corresponding flux according to the wild type flux range and the overexpression type flux range; reconstructing a reaction equation by using an atomic mapping tool for the operation to be performed, obtaining main substrates and products in a reaction corresponding to the operation, and constructing a metabolic network connectivity graph according to the reaction occurrence direction; adding the corresponding flux as a weight value to the metabolic network connectivity graph; adding a virtual starting node as a starting point of all nodes with an in-degree of 0, adding a virtual ending node as an ending point of all nodes with an out-degree of 0, and solving a key site; and selecting a preferentially regulated reaction for the key site.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of biological metabolism, and in particular to a biological metabolic pathway regulation method and device. BACKGROUND

[0002] Flux balance analysis based on mass and charge conservation has been proven to guide metabolic engineering strategies to improve target product yield, but the flux balance analysis algorithm gives only one set of solutions, and it is difficult to achieve all reaction fluxes in experiments to match the calculations, while the flux difference analysis algorithm can give the range of each reaction flux under the optimal production product condition, which is more convenient for experimental personnel to analyze and adjust, but flux difference analysis often gives a large number of reactions that need to be modified, which is difficult to verify one by one. In order to facilitate the application of flux difference analysis algorithm, researchers have provided some algorithms, such as optknock algorithm, optforce algorithm, etc., which can provide modification suggestions for experimental personnel to choose, but these algorithms are mostly based on mathematical optimization equations. If a metabolic map can be constructed, according to the upstream and downstream logical relationship of the metabolic map, it is suggested to preferentially regulate upstream reactions, which may affect downstream reactions. The reactions that need to be preferentially regulated often need to consider the flux size of the reaction under wild-type conditions, and the reactions on the same metabolic pathway are usually preferentially regulated to the reactions with larger flux difference between the overexpression type and the wild-type reaction, which may have more obvious regulation effect. Therefore, a more practical metabolic pathway regulation method is needed. SUMMARY

[0003] In view of this, the present application provides a biological metabolic pathway regulation method and device.

[0004] According to one aspect of the present application, a biological metabolic pathway regulation method is provided, which comprises:

[0005] Solving the wild-type flux range and the overexpression type flux range of the metabolic network model;

[0006] Determining the direction of reaction occurrence according to the overexpression type flux range;

[0007] Determining the operation to be performed and the corresponding flux according to the wild-type flux range and the overexpression type flux range;

[0008] Using an atomic mapping tool to reconstruct the reaction equation for the operation to be performed, obtaining the main substrate and product in the reaction corresponding to the operation, and constructing a metabolic network connectivity graph according to the direction of reaction occurrence;

[0009] Adding the corresponding flux as a weight to the metabolic network connectivity graph;

[0010] Adding a virtual starting node as the starting point of all nodes with in-degree of 0, adding a virtual ending node as the ending point of all nodes with out-degree of 0, and solving the key sites;

[0011] Selecting a reaction to be preferentially regulated for the key site.

[0012] Preferably, the method further comprises: adding necessary exogenous reactions to the metabolic network model to enable the metabolic network model to reach the product from the substrate.

[0013] Preferably, the method further comprises: drawing a metabolic flow chart after selecting the reaction to be preferentially regulated.

[0014] Preferably, the wild-type flux absolute value minimum is greater than the overexpression-type flux absolute value maximum, the operation to be performed is down-regulation, and the corresponding flux is the difference between the wild-type flux absolute value minimum and the overexpression-type flux absolute value maximum; the wild-type flux absolute value maximum is less than the overexpression-type flux absolute value minimum, the operation to be performed is up-regulation, and the corresponding flux is the difference between the overexpression-type flux absolute value minimum and the wild-type flux absolute value maximum; the overexpression-type flux is 0, the operation to be performed is knockout, and the corresponding flux is the smaller one of the absolute value of the wild-type flux maximum and the absolute value of the wild-type flux minimum.

[0015] Preferably, the metabolic network connectivity graph removes the cofactors in the reaction equation and removes the small molecule substances.

[0016] Preferably, constructing the metabolic network connectivity graph comprises: constructing the metabolic network connectivity graph based on the reconstructed reaction equation and in combination with the direction of reaction occurrence.

[0017] Preferably, solving the key site comprises: solving the key site in the metabolic network connectivity graph by using the maximum flow minimum cut algorithm.

[0018] Preferably, selecting the reaction to be preferentially regulated for the key site comprises: counting the flux absolute values of the multi-step reactions contained in the key site, and selecting the reaction with the maximum flux absolute value to be preferentially regulated.

[0019] The application provides a biological metabolic pathway regulation device, which comprises a processor and a memory, the memory stores a computer program, and the processor executes the computer program to realize the method.

[0020] According to the technical solution of the present application, the existing biological metabolism calculation method is combined with the graph theory of a computer, important atoms participating in reactions are found out through atom mapping, the purpose of removing small molecule substances and cofactors to construct a more reasonable connected graph is achieved, the minimum intervention set is given using the maximum flow minimum cut algorithm, and a reasonable operation site is found out, so that the production effect of the maximized target product under the condition of the minimum intervention is achieved. For the knockout, up-regulation and down-regulation reaction set, the key reactions of gene operation can be given, and the transformation cost, time and energy of experimental personnel can be greatly saved by transforming these key reactions.

[0021] Other features and advantages of the present application will be described in detail in the following specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0022] The accompanying drawings, which form a part of the present application, are intended to provide further understanding of the present application, and the illustrative embodiments of the present application and their description serve the explanation of the present application. In the drawings:

[0023] Figure 1 The biological metabolic pathway regulation method provided in the present application;

[0024] Figure 2 The reaction flux ranking result schematic diagram;

[0025] Figure 3 The regulation result schematic diagram provided in the present application;

[0026] Figure 4 The overall flow of key site selection provided in the present application;

[0027] Figure 5 The flux range solving and regulation reaction division explanation schematic diagram provided in the present application;

[0028] Figure 6 The reactant id and its name in the reaction equation provided in the present application;

[0029] Figure 7 The reaction equation provided in the present application. DETAILED DESCRIPTION

[0030] It should be noted that the embodiments in the present application and the features in each embodiment can be combined with each other without conflict.

[0031] The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0032] In the present application, the biological metabolic pathway is regulated based on the metabolic network model and combined with the maximum flow minimum cut algorithm. The biological metabolic pathway regulation method provided in the present application, as shown in Figure 1 specifically includes:

[0033] Step 101, using flux variability analysis algorithm to solve the wild type reaction flux range of each reaction under the condition of maximum growth as objective function for the metabolic network model. The minimum reaction flux absolute value is recorded as wild_min_flux, and the maximum reaction flux absolute value is recorded as wild_max_flux.

[0034] Step 102, using flux variability analysis algorithm to solve the overexpression flux range of each reaction under the condition of product production as objective function for the metabolic network model. The minimum reaction flux absolute value is recorded as express_min_flux, and the maximum reaction flux absolute value is recorded as express_max_flux.

[0035] Step 103, determine the direction of reaction according to the overexpression flux range. For example, there is a reaction equation The reaction can occur in both directions, and the upper limit of flux is defined to be greater than or equal to the lower limit of flux. If the overexpression flux lower limit is greater than zero, it means that the reaction should occur in the forward direction, and the overexpression flux upper limit is less than zero, which means that the reaction should occur in the reverse direction.

[0036] Step 104, compare the wild type and overexpression reaction flux range, and determine whether up-regulation, down-regulation or knockout operation is needed according to whether there is an intersection of the range.

[0037] Step 105, record the up-regulation and down-regulation control interval respectively, that is, how much wild type up-regulation and down-regulation flux can achieve the flux requirement of overexpression. Take the up-regulation reaction as an example, record it as upgrade_flux=express_min_flux-wild_max_flux. For knockout, select the flux with the smallest absolute value of the upper and lower limits of the wild type reaction flux, and the absolute value of this flux is the flux that needs to be adjusted for knockout operation.

[0038] Step 106, process the up-regulation, down-regulation and knockout reactions respectively. Take the up-regulation reaction as an example, use the atom mapping tool for each step of the reaction, and obtain the main substrate and product according to the transfer of atomic groups. Remove the cofactors in the reaction equation and remove small molecule substances for the purpose of constructing a reasonable metabolic network graph.

[0039] Step 107, construct a metabolic network connectivity graph based on the main reaction equation obtained in the last step.

[0040] Step 108, add the flux that needs to be up-regulated as the weight value for the reactions in the metabolic network graph, that is, the above upgrade_flux.

[0041] Step 109, add a virtual starting node as the starting point of all metabolites with in-degree of 0, denoted as start_node, and add a virtual end node as the end point of all nodes with out-degree of 0, denoted as end_node.

[0042] Step 110, apply the maximum flow minimum cut algorithm to the modified metabolic network graph to solve the key sites.

[0043] Step 111, the key sites often contain multi-step reactions, count the absolute value of the flux of the multi-step reactions, and select the reaction with the largest absolute value of the flux for priority control, that is, select the reaction with the largest absolute value of the flux for priority control.

[0044] Preferably, before implementing the above method, the metabolic network model can also be modified to add necessary exogenous reactions to enable the substrate to reach the product.

[0045] Preferably, after determining the priority control reaction, a metabolic flow chart can be drawn.

[0046] The following is an example of Escherichia coli iML1515.mat model. Since Escherichia coli iML1515.mat model itself can produce lysine, it does not need to add exogenous reactions. The control process includes:

[0047] Step 201, use the flux difference analysis algorithm to obtain the flux of each reaction under the condition of maximum growth, wherein the lower limit of the absolute value of the flux is recorded in the list wild_min_flux, and the upper limit of the absolute value of the flux is recorded in the list wild_max_flux.

[0048] Step 202, use the flux difference analysis algorithm to obtain the flux of each reaction under the condition of producing product as the objective function, wherein the lower limit of the absolute value of the flux is recorded in the list express_min_flux, and the upper limit of the absolute value of the flux is recorded in the list express_max_flux.

[0049] Step 203, record the reaction direction according to the overexpression flux range.

[0050] Step 204, taking up-regulation as an example. Calculate the flux difference interval of each reaction upgrade_flux=express_min_flux-wild_max_flux as the weight for subsequent construction of connected graph.

[0051] Step 205, reconstruct the reaction equation based on the atom mapping rule. Get the metabolites and products of each reaction core.

[0052] Step 205, based on the reconstructed reaction equation, construct a connected graph combined with the reaction direction.

[0053] Step 206, adding weights to the connected graph.

[0054] Step 207, adding a virtual start node as the starting point of all metabolites with an in-degree of 0, denoted as start node, and a virtual end node as the end point of all nodes with an out-degree of 0, denoted as end node.

[0055] Step 208, applying the maximum flow minimum cut algorithm to solve the operation site.

[0056] Step 209, sorting the regulated reaction fluxes to obtain the preferentially regulated reactions. Figure 2 The reaction set (first column), total flux sum (second column), regulation mode (third column), preferentially operated reactions (fourth column), and preferentially operated reaction fluxes (fifth column) are shown. For example, the third row requires knocking out the PTAr, PDH, and AKGDH three-step reactions, and the total flux of these three reactions is 16.17032501558058, of which the PDH reaction flux is 9.68968679340153, accounting for the highest proportion, and this step is preferentially regulated.

[0057] Figure 3 The shown regulation result shows that strengthening the upstream reaction G6PDH2r can improve the yield of lysine, which is also confirmed in the literature. Other reactions are the same.

[0058] Figure 4 The overall flow of selecting key sites provided by the present application is shown, which specifically includes:

[0059] Step 401, solving the flux range;

[0060] Step 402, classifying the reactions to be regulated into three categories: up-regulation, down-regulation, and knockout;

[0061] Step 403, reconstructing the metabolic reaction equation based on atomic mapping;

[0062] Step 404, constructing a connected graph for each category;

[0063] Step 405, using the maximum flow minimum cut algorithm to obtain the key site for each connected graph.

[0064] Figure 5 The illustration of solving the flux range and classifying the regulated reactions is shown, wherein the minimum absolute value of the wild-type flux is greater than the maximum absolute value of the overexpression-type flux, down-regulation is performed; the maximum absolute value of the wild-type flux is less than the minimum absolute value of the overexpression-type flux, up-regulation is performed; the wild-type flux absolute value range and the overexpression-type flux absolute value range overlap, no regulation is performed; and the overexpression-type flux is 0, then knockout is performed.

[0065] In order to realize the technical scheme of the present application, it is necessary to reconstruct the reaction equation based on atomic mapping. The present application constructs a logically connected graph according to metabolites for a single regulation reaction set, and then gives the reaction that needs to be regulated for each connected graph. Since there are many small molecules in biochemical reactions, such as h2o, co2, etc., if these small molecules are used as a bridge to connect two reactions, it is unreasonable to construct a connected graph in this logical relationship; those skilled in the art can determine the small molecules in advance. For example, for the three reactions recorded in BIGG: ASPK: asp__L_c + atp_c <=> adp_c + 4pasp_c, ASAD: aspsa_c + pi_c + nadp_c <=> h_c + nadph_c + 4pasp_c, DHDPS: aspsa_c + pyr_c -> h_c + 2h2o_c + 23dhdp_c. They are reactions related to lysine synthesis path. If the upstream and downstream relationships of the three reactions are to be constructed, it should be that the reaction ASPK occurs first, 4pasp is generated from asp__L, then the reaction ASAD occurs, 4pasp generates aspsa, and aspsa is used as an intermediate product to connect the reactions ASAD and DHDPS, and aspsa generates 23dhdp. It is more reasonable to connect the three reactions through the two core intermediate products. In reactions ASAD and DHDPS, there is also a common metabolite, which is h. Generally, small molecule metabolites such as h2o, h, atp are not used as intermediate products to connect multiple reactions, because many reactions need the participation of h2o, h, atp, etc. If these substances are used as intermediate products to connect metabolic reactions, the constructed metabolic reactions cannot play their due role. Figure 6 The id of the reactant and its name are shown.

[0066] By using the rdt toolkit, the reaction equation is constructed as a single substrate to product reaction equation. The rdt toolkit can perform atomic mapping on biochemical reactions. Even in unbalanced reactions (the number of atoms on both sides of the reaction equation is different), reasonable guesses will be made. For reaction DHDPS: aspsa_c + pyr_c -> h_c + 2h2o_c + 23dhdp_c, the main substrate obtained is M00002, that is, aspsa atomic transfer accounts for a high proportion, and the main product M00003 is 23dhdp, excluding other metabolites that contribute less in the reaction, the reaction equation is reconstructed, and the reaction DHDPS is reconstructed as aspsa -> 23dhdp. Figure 7 The reaction formula is shown. M00001 represents pyr, M00004 represents h2o, and M00005 represents h, which is not the main product.

[0067] Correspondingly, the application provides a biological metabolic pathway regulation device, which comprises a processor and a memory, the memory stores a computer program, and the processor executes the computer program to realize the method of the application.

[0068] The application provides a biological metabolic pathway regulation method and device. The method comprises: solving a wild-type flux range and an overexpression-type flux range of a metabolic network model; determining a reaction occurrence direction according to the overexpression-type flux range; determining an operation to be performed and a corresponding flux according to the wild-type flux range and the overexpression-type flux range; reconstructing a reaction equation using an atom mapping tool for the operation to be performed, obtaining main substrates and products in a reaction corresponding to the operation, and constructing a metabolic network connectivity graph according to the reaction occurrence direction; adding the corresponding flux as a weight to the metabolic network connectivity graph; adding a virtual starting node as a starting point of all nodes with an in-degree of 0 and adding a virtual ending node as an ending point of all nodes with an out-degree of 0, and solving a key site; and selecting a preferentially regulated reaction for the key site. According to the technical solution of the application, an existing biological metabolic calculation method is combined with graph theory of a computer, important atoms participating in a reaction are solved through atom mapping, and a more reasonable connectivity graph is constructed.

[0069] The above only describes preferred embodiments of the application and is not intended to limit the application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the application shall be included in the protection scope of the application.

Claims

1. A method for regulating biological metabolic pathways, characterized in that, The method comprises: Solving the wild-type flux range and the overexpression-type flux range of the metabolic network model; Determining the direction of reaction occurrence according to the overexpression-type flux range; Determining the operation to be performed and the corresponding flux according to the wild-type flux range and the overexpression-type flux range; Reconstructing the reaction equation using the atomic mapping tool for the operation to be performed, obtaining the main substrate and product in the reaction corresponding to the operation, and constructing a metabolic network connectivity graph according to the direction of reaction occurrence; Adding the corresponding flux as a weight to the metabolic network connectivity graph; Adding a virtual starting node as the starting point of all nodes with an in-degree of 0, adding a virtual ending node as the ending point of all nodes with an out-degree of 0, and solving the critical site; Selecting the reaction to be preferentially regulated for the critical site; Solving the wild-type reaction flux range of the metabolic network model using the flux difference analysis algorithm with the maximum growth as the objective function; Solving the overexpression-type flux range of each reaction of the metabolic network model using the flux difference analysis algorithm with the production of the product as the objective function; The minimum value of the wild-type flux absolute value is greater than the maximum value of the overexpression-type flux absolute value, the operation to be performed is down-regulation, and the corresponding flux is the difference between the minimum value of the wild-type flux absolute value and the maximum value of the overexpression-type flux absolute value; the maximum value of the wild-type flux absolute value is less than the minimum value of the overexpression-type flux absolute value, the operation to be performed is up-regulation, and the corresponding flux is the difference between the minimum value of the overexpression-type flux absolute value and the maximum value of the wild-type flux absolute value; the overexpression-type flux is 0, the operation to be performed is knockout, and the corresponding flux is the smaller one of the absolute value of the maximum value and the absolute value of the minimum value of the wild-type flux; The metabolic network connectivity graph removes the cofactors in the reaction equation and removes the small molecule substances; Solving the critical site includes using the maximum flow minimum cut algorithm to solve the critical site in the metabolic network connectivity graph.

2. The method of claim 1, wherein, The method further comprises adding necessary exogenous reactions to the metabolic network model to enable the metabolic network model to reach the product from the substrate.

3. The method of claim 1, wherein, The method further comprises drawing a metabolic flow chart after selecting the reaction to be preferentially regulated.

4. The method of claim 1, wherein, Constructing the metabolic network connectivity graph comprises constructing the metabolic network connectivity graph based on the reconstructed reaction equation and combining the direction of reaction occurrence.

5. The method of claim 1, wherein, Selecting the reaction to be preferentially regulated for the critical site comprises: counting the flux absolute value of the multi-step reaction contained in the critical site, and selecting the reaction with the maximum flux absolute value for preferential regulation.

6. A biological metabolic pathway regulation apparatus comprising a processor and a memory, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the method according to any one of claims 1-5.

Images