Metabolic state determination method and device and computer readable storage medium

Abstract

The invention provides a metabolic state determination method and device and a computer readable storage medium, and the method comprises the steps: carrying out the preprocessing of a metabonomics missing data matrix through employing normal distribution, and obtaining a preprocessed data matrix; performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix; updating the missing data matrix according to the diagonal matrix to obtain a first approximate matrix; interpolating missing data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix; and repeating from the process of taking the second approximate matrix as a preprocessing data matrix to carry out singular value decomposition until an iteration condition is reached, taking the finally obtained second approximate matrix as a final interpolation result, and determining the metabolic state of the living body according to the interpolated metabonomics data. According to the method, the missing data is interpolated by selecting the corresponding interpolation mode according to the missing type, so that the application scenarios of the metabolic state determination method are increased.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular, to a method and device for determining a metabolic state, and a computer-readable storage medium. Background Art

[0002] Due to technical limitations, errors in sample collection and processing, and insufficient sensitivity and coverage of metabolite detection technology, the lack of metabolomics data has become a common problem. The lack of metabolomics data directly leads to a decrease in accuracy when analyzing the metabolic state of an organism. The current common solution is to interpolate the missing metabolomics data to determine the metabolic state of the organism.

[0003] Traditional methods for determining metabolic status mainly include mean interpolation, median interpolation, mode interpolation, regression interpolation, K nearest neighbor interpolation (KNN), and multiple interpolation. These methods have their own advantages and disadvantages and are suitable for different data types and missing patterns. For example, mean or median interpolation is simple and fast, but may not be accurate enough for non-random missing or highly skewed data; regression interpolation considers the relationship between variables, but assumes that data missing is random; KNN interpolation is based on the principle of similarity and is suitable for more complex data structures, but the amount of calculation is large; multiple interpolation reflects the uncertainty of interpolation by generating multiple complete data sets, which is a more advanced method.

[0004] Although there are a variety of data interpolation techniques, most existing methods are designed for a single type of missing data, such as only dealing with random missing, completely random missing, or a specific type of non-random missing. This limitation makes it difficult for them to cope with complex and changing practical application scenarios, which in turn affects the accuracy of metabolic data and leads to low accuracy in determining the metabolic state of an organism. Summary of the invention

[0005] In view of this, the purpose of the embodiments of the present application is to provide a metabolic state determination method, device and computer-readable storage medium, which can improve the accuracy of metabolomics and the metabolic state of an organism, and at the same time increase the application scenarios of the metabolic state determination method.

[0006] In a first aspect, an embodiment of the present application provides a method for determining a metabolic state, comprising: preprocessing a missing data matrix of metabolomics using a normal distribution to obtain a preprocessed data matrix; wherein the metabolomics is configured to reveal the metabolic state of an organism under different physiological or pathological states; performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix; updating the missing data matrix according to the diagonal matrix to obtain a first approximate matrix; interpolating data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix; repeating from performing singular value decomposition on the second approximate matrix as the preprocessed data matrix until an iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result; and determining the metabolic state of the organism based on the interpolated metabolomics data.

[0007] In the above implementation process, when interpolating data, the data in the first approximate matrix are interpolated together according to the missing data matrix, the preprocessed data matrix and the first approximate matrix, wherein the preprocessed data matrix can be used to determine the missing type, so that the corresponding interpolation method can be selected according to the missing type to interpolate the missing data during interpolation, which can improve the accuracy of metabolomics and the metabolic state of the organism, and increase the application scenarios of the metabolic state determination method. In addition, since the singular value decomposition method has certain advantages in matrix dimension reduction and data reconstruction, in the method of data interpolation, the singular value decomposition method is used for interpolation, which can reduce the difficulty of interpolation and improve the accuracy of interpolation. Furthermore, since the normal distribution only depends on the two parameters of mean and variance, by using the normal distribution to preprocess the missing data matrix of metabolomics, the processing difficulty of the missing data matrix can be simplified and the processing efficiency can be improved.

[0008] In one embodiment, the diagonal matrix includes a first diagonal matrix and a second diagonal matrix; performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix includes: performing singular value decomposition on the preprocessed data matrix to obtain a first diagonal matrix containing singular values; processing the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix; updating the missing data matrix according to the diagonal matrix to obtain a first approximate matrix includes: updating the missing data matrix according to the second diagonal matrix to obtain a first approximate matrix.

[0009] In the above implementation process, by performing singular value decomposition on the preprocessed data matrix, high-dimensional data can be reduced to low-dimensional data, effectively reducing the dimension of the data and reducing the computational complexity. In addition, by retaining the features corresponding to the singular values, important information in the preprocessed data matrix can be retained, improving the accuracy of data interpolation.

[0010] In one embodiment, the method for obtaining the first diagonal matrix includes: using the PX-MDC model to classify the missing data to obtain the missing type; when the missing type of the missing data is a non-random missing type, randomly weighting the preprocessed data of the non-random missing type to obtain a randomly weighted preprocessed data matrix; performing singular value decomposition on the randomly weighted preprocessed data matrix to obtain a first diagonal matrix containing singular values.

[0011] In the above implementation process, when the missing type is non-random missing, by weighting the data before interpolation to give higher weight to the observed data, reliable data can be used to the greatest extent, while the weight of the non-random missing type data can be reduced, reducing the negative impact of missing data on matrix decomposition and improving the flexibility and robustness of the model.

[0012] In one embodiment, the processing of the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix includes: subtracting the set threshold from all singular values ​​in the first diagonal matrix to obtain an intermediate matrix; retaining a first preset proportion of data in the intermediate matrix to obtain the second diagonal matrix; wherein the first preset proportion is the percentage of the sum of the retained singular values ​​to the total singular values.

[0013] In the above implementation process, after determining the first diagonal matrix, by setting the first preset ratio, the data of the first preset ratio in the intermediate matrix is ​​retained, and by adjusting the value of the first preset ratio, as many singular values ​​as possible can be retained to optimize the interpolation effect.

[0014] In one embodiment, the use of normal distribution to preprocess the missing data matrix of metabolomics to obtain a preprocessed data matrix includes: replacing missing values ​​with the minimum value of the metabolites in the original data that are not missing; calculating the mean and variance of the metabolites after replacing the missing values; generating normally distributed data based on the mean and the variance; and performing corresponding preprocessing on the normally distributed data according to the missing type of the missing data to obtain a preprocessed data matrix.

[0015] In the above implementation process, by performing normal distribution processing on the missing data matrix, the data distribution characteristics in the missing data matrix are made more intuitive and easy to explain, the difficulty of processing the missing data matrix is ​​simplified, and the convenience of processing the missing data matrix is ​​improved.

[0016] In one embodiment, the normally distributed data is preprocessed accordingly according to the missing type of the missing data to obtain a preprocessed data matrix, including: when the missing data is a randomly missing type, the missing data is filled by randomly selected values ​​in the normally distributed data; when the missing data is a non-random missing type, data less than the minimum value of the metabolite is selected from the normally distributed data to fill the missing data; and the preprocessed data matrix is ​​obtained through the missing data matrix after filling.

[0017] In the above implementation process, when preprocessing the missing data matrix, the missing data is filled in using a corresponding filling method according to the missing type of the missing data, which can improve the filling accuracy.

[0018] In one embodiment, the data in the first approximate matrix is ​​interpolated according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix, including: determining the missing position of the missing data according to the missing data matrix; determining the missing type of the missing data according to the preprocessed data matrix; determining the missing position and missing type of the missing data in the first approximate matrix by the missing position and missing type of the missing data; interpolating the corresponding missing positions in the first approximate matrix using a corresponding interpolation method according to the missing type of the missing data in the first approximate matrix to determine the second approximate matrix.

[0019] In the above implementation process, when determining the second approximate matrix, the missing position of the missing data is determined according to the missing data matrix, and the missing type of the missing data is determined according to the preprocessed data matrix, and then the corresponding interpolation method is determined to interpolate at the corresponding processing position of the first approximate matrix. The interpolation data can be accurately interpolated to the corresponding position, thereby improving the accuracy of the second approximate matrix.

[0020] In one embodiment, the second approximate matrix is ​​determined by interpolating the corresponding missing positions in the first approximate matrix using the corresponding interpolation method according to the missing type of the missing data in the first approximate matrix, including: when the missing data is a random missing type, the missing data is filled by randomly selected values ​​in the normally distributed data; when the missing data is a non-random missing type, and the value of the missing data position of the missing data at the corresponding position in the first approximate matrix is ​​higher than the minimum value of the metabolite in the original data that is not missing, the missing data is filled by the first weight value of the corresponding position of the preprocessed data matrix and the second weight value of the corresponding position in the first approximate matrix; when the missing data is a non-random missing type, and the value of the missing data at the corresponding position in the first approximate matrix is ​​not higher than the minimum value of the metabolite in the original data that is not missing, the missing data is filled by the value of the corresponding position in the first approximate matrix.

[0021] In the above implementation process, for missing data of different missing types, corresponding interpolation methods are used to interpolate missing data, which can improve the accuracy of missing data interpolation, increase the missing types that can be targeted by the data interpolation method, and increase the application scenarios.

[0022] In a second aspect, an embodiment of the present application further provides a metabolic state determination device, comprising: a preprocessing module, used to preprocess the missing data matrix of metabolomics using a normal distribution to obtain a preprocessed data matrix; wherein the metabolomics is configured to reveal the metabolic state of an organism under different physiological or pathological states; a decomposition module, used to perform singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix; an update module, used to update the missing data matrix according to the diagonal matrix to obtain a first approximate matrix; an interpolation module, used to interpolate the data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix; an iteration module, used to repeat starting from performing singular value decomposition on the second approximate matrix as the preprocessed data matrix until an iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result; a determination module, used to determine the metabolic state of the organism based on the interpolated metabolomics data.

[0023] In a third aspect, an embodiment of the present application further provides an electronic device, comprising: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the machine-readable instructions are executed by the processor to perform the steps of the method in the above-mentioned first aspect, or any possible implementation of the first aspect.

[0024] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the metabolic state determination method in the above-mentioned first aspect, or any possible implementation of the first aspect are executed.

[0025] In order to make the above-mentioned objects, features and advantages of the present application more obvious and understandable, embodiments are given below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0027] Figure 1 A block diagram of an electronic device provided in an embodiment of the present application; Figure 2 A flow chart of a method for determining a metabolic state provided in an embodiment of the present application; Figure 3 NRMSE of different interpolation methods on the ST000419 dataset in the experiment provided in the embodiment of the present application; Figure 4 NRMSE of different interpolation methods on the ST000118 dataset in the experiment provided in the embodiment of the present application; Figure 5 SOR of different interpolation methods on the ST000419 data set in the experiment provided in the embodiment of the present application; Figure 6 SOR of different interpolation methods on the ST000118 data set in the experiment provided in the embodiment of the present application; Figure 7 This is a schematic diagram of the functional modules of the metabolic state determination device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings in the embodiments of the present application.

[0029] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0030] Missing values ​​in metabolomics data are caused by a variety of technical and biological factors, such as missing low-concentration metabolites due to insufficient instrument detection sensitivity, randomly introduced missing values, and data loss caused by changes in the measurement environment. Based on these factors, missing values ​​can be divided into missing types including completely random missing values, random missing values, and non-random missing values. Completely random missing values ​​refer to missing values ​​whose probability is independent of any observed or unobserved values ​​and are therefore considered completely random. Random missing values ​​occur when the probability of missing values ​​depends on the observed data, such as those caused by technical problems such as inaccurate peak detection, signal overlap deconvolution, or ion suppression. Non-random missing values ​​are missing values ​​caused by metabolite concentrations below the detection limit, and are usually unevenly distributed and concentrated on specific metabolites.

[0031] In order to solve the problem of missing data, technicians have proposed a series of data interpolation methods. For example, for completely random missing data, commonly used interpolation methods include random forest interpolation, K nearest neighbor interpolation, singular value decomposition interpolation, and multiple interpolation chain equations.

[0032] Among them, random forest interpolation uses multiple decision trees to capture complex relationships in the data. It is suitable for high-dimensional data and is robust to outliers. K-nearest neighbor interpolation finds the k neighbors that are most similar to the missing sample and fills in the missing value based on the values ​​of these neighbors. The method is simple and intuitive, but the computational cost is high in high-dimensional data. Singular value decomposition interpolation captures the main structure of the data through matrix decomposition technology, which can effectively reduce dimensionality and noise, but relies on the linear assumption of the data. Multiple interpolation chain equations reflect the uncertainty of missing values ​​by generating multiple interpolation data sets, providing more reliable statistical inferences, but the computational complexity is high.

[0033] The processing of non-random missing data is more complicated. Commonly used methods include semi-minimum interpolation, quantile regression interpolation for left-deleted missing data, and No-Skip K nearest neighbor interpolation. Among them, semi-minimum interpolation is a method to infer missing values ​​based on assumptions about the data, which can be estimated by the statistical characteristics of existing data. However, this method may lead to serious deviations when the assumptions are not true, especially when the data distribution is uneven, the accuracy of the interpolation results will be affected. The quantile regression interpolation method for left-deleted missing data can effectively handle the interpolation of left-censored data by constructing a quantile regression model, especially for cases where the data distribution is asymmetric. The advantage of this method is that it is more robust to extreme values, but it is highly dependent on the model setting. It is necessary to ensure that the selected quantile is suitable for the data characteristics, and the computational complexity is relatively high. No-Skip K nearest neighbor interpolation avoids the information loss in the traditional KNN method by retaining all available data. Its main advantage is that it can use more information for interpolation, which enhances the accuracy of interpolation.

[0034] However, the inventors of this application have found through long-term research that the actual missing data of metabolomics are more complicated, and there are usually mixed missing data of various types, such as completely random missing data, random missing data, and non-random missing data. Therefore, the above-mentioned traditional single strategies are difficult to deal with effectively, which in turn affects the accuracy of metabolic data and leads to low accuracy in determining the metabolic state of an organism.

[0035] In view of this, the present application proposes a method for determining metabolic state, when interpolating data, interpolating the data in the first approximate matrix according to the missing data matrix, the pre-processed data matrix and the first approximate matrix, wherein the pre-processed data matrix can be used to determine the missing type, so that the corresponding interpolation method can be selected according to the missing type to interpolate the missing data during interpolation, which can improve the accuracy of metabolomics and the metabolic state of the organism, and increase the application scenarios of the method for determining the metabolic state. In addition, since the singular value decomposition method has certain advantages in matrix dimension reduction and data reconstruction, in the method for data interpolation, interpolation is performed by using the singular value decomposition method, which can reduce the difficulty of interpolation and improve the accuracy of interpolation. Furthermore, since the normal distribution only depends on the two parameters of mean and variance, by pre-processing the missing data matrix of metabolomics using the normal distribution, the processing difficulty of the missing data matrix can be simplified and the processing efficiency can be improved.

[0036] To facilitate understanding of this embodiment, first, an electronic device that executes the metabolic state determination method disclosed in the embodiment of the present application is introduced in detail.

[0037] like Figure 1 , which is a block diagram of an electronic device. The electronic device 100 may include a memory 111 and a processor 113. A person skilled in the art may understand that Figure 1 The structure shown is only for illustration and does not limit the structure of the electronic device 100. For example, the electronic device 100 may further include Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0038] The memory 111 and the processor 113 are directly or indirectly electrically connected to each other to achieve data transmission or interaction. For example, these elements can be electrically connected to each other via one or more communication buses or signal lines. The processor 113 is used to execute the executable module stored in the memory.

[0039] The memory 111 may be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc. The memory 111 is used to store programs, and the processor 113 executes the program after receiving an execution instruction. The method executed by the electronic device 100 defined by the process disclosed in any embodiment of the present application can be applied to the processor 113, or implemented by the processor 113.

[0040] The processor 113 may be an integrated circuit chip with signal processing capability. The processor 113 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0041] The electronic device 100 in this embodiment can be used to execute each step in each method provided in the embodiments of the present application. The implementation process of the metabolic state determination method is described in detail below through several embodiments.

[0042] See also Figure 2 , is a flow chart of the method for determining metabolic state provided in the embodiment of the present application. Figure 2 The specific process shown is described in detail.

[0043] Step 201 : preprocessing the missing data matrix of metabolomics using normal distribution to obtain a preprocessed data matrix.

[0044] Among them, metabolomics is a new discipline that simultaneously conducts qualitative and quantitative analysis of all low-molecular-weight metabolites in a certain organism or cell during a specific physiological period. By analyzing and interpreting the metabolite spectra in biological samples, the metabolic state of the organism under different physiological or pathological states can be revealed.

[0045] The normal distribution mentioned above is a probability distribution, which is used to describe the probability distribution of the values ​​of a random variable within a certain range.

[0046] Understandably, in real data, metabolites may not follow a normal distribution. To improve this situation, the data can be log-transformed to make it close to a normal distribution. After completing the data interpolation, an exponential transformation is applied to restore the data to its original proportion, ensuring that the interpolation process does not change the essential distribution characteristics of the data.

[0047] Step 202: Perform singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix.

[0048] The diagonal matrix may include a first diagonal matrix and a second diagonal matrix. The first diagonal matrix is ​​a matrix obtained by performing singular value decomposition on the preprocessed data matrix, and the second diagonal matrix is ​​a matrix obtained by processing the first diagonal matrix.

[0049] It can be understood that the diagonal matrix after singular value decomposition may contain singular values.

[0050] Step 203: Update the missing data matrix according to the diagonal matrix to obtain a first approximate matrix.

[0051] The first approximate matrix here can be further inferred based on the singular value decomposition formula.

[0052] It should be understood that after the preprocessed data matrix is ​​decomposed by singular value, a diagonal matrix is ​​obtained. By further processing the diagonal matrix, a processed diagonal matrix (i.e., a second diagonal matrix) may be obtained. Substituting the processed diagonal matrix into the above singular value decomposition formula, a new preprocessed data matrix (i.e., a first approximate matrix) can be obtained.

[0053] Step 204: interpolate the data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix.

[0054] Among them, the missing data matrix is ​​the original metabolomics data with missing data, and the missing data matrix can be used to determine the missing position of the missing data. The preprocessed data matrix refers to the matrix after the missing data matrix has undergone simple data processing, and the preprocessed data matrix can be used to determine the missing type of the missing data in the missing data matrix. The first approximate matrix is ​​the missing data matrix obtained after processing the diagonal matrix after singular value decomposition.

[0055] It can be understood that when interpolating the data in the first approximate matrix, a corresponding interpolation method can be determined according to the missing type of the missing data to perform the interpolation.

[0056] The missing data of metabolomics here include completely random missing, random missing and non-random missing.

[0057] Among them, the data interpolation methods corresponding to different missing types can be different.

[0058] Step 205, starting from taking the second approximate matrix as the preprocessed data matrix to perform singular value decomposition, repeating until the iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result.

[0059] It should be understood that after the second approximate matrix is ​​determined, the second approximate matrix is ​​used as a preprocessing data matrix to continue to perform singular value decomposition to obtain a new diagonal matrix, and then a new first approximate matrix is ​​obtained based on the new diagonal matrix, and the new first approximate matrix is ​​interpolated to obtain a new second diagonal matrix. The new second diagonal matrix is ​​then used as a preprocessing data matrix to continue to perform singular value decomposition... Repeated until the iteration condition is met, and the final second approximate matrix is ​​the final interpolation result.

[0060] The iteration condition here may be reaching a convergence condition, or reaching a maximum number of iterations, etc. The iteration condition may be selected according to actual conditions.

[0061] Step 206, determine the metabolic state of the organism according to the metabolomics data after interpolation. In the above implementation process, when interpolating data, interpolate the data in the first approximate matrix according to the missing data matrix, the pre-processed data matrix and the first approximate matrix, wherein the pre-processed data matrix can be used to determine the missing type, so that the corresponding interpolation method can be selected according to the missing type to interpolate the missing data during interpolation, which can improve the accuracy of metabolomics and the metabolic state of the organism, and increase the application scenarios of the metabolic state determination method. In addition, since the singular value decomposition method has certain advantages in matrix dimension reduction and data reconstruction, in the method for data interpolation, interpolation is performed by using the singular value decomposition method, which can reduce the difficulty of interpolation and improve the accuracy of interpolation. Furthermore, since the normal distribution only depends on the two parameters of mean and variance, by pre-processing the missing data matrix of metabolomics using the normal distribution, the processing difficulty of the missing data matrix can be simplified and the processing efficiency can be improved.

[0062] In a possible implementation, step 202 includes: performing singular value decomposition on the preprocessed data matrix to obtain a first diagonal matrix containing singular values; and processing the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix.

[0063] In one embodiment, the formula for singular value decomposition may be: ; in, is the preprocessed data matrix, and is an orthogonal matrix, is the first diagonal matrix.

[0064] The threshold value here may be a value set in advance. For example, the threshold value may be 0, 1, 5, etc. The threshold value may be adjusted according to actual conditions.

[0065] It can be understood that after the singular value decomposition is performed, a new diagonal matrix (ie, the second diagonal matrix) can be re-determined according to the singular values ​​in the matrix obtained after the singular value decomposition and a set threshold.

[0066] In one embodiment, step 203 includes: updating the missing data matrix according to the second diagonal matrix to obtain a first approximate matrix.

[0067] Specifically, the first approximate matrix can be obtained by the following formula: ; in, is the first approximate matrix, and is an orthogonal matrix, is the second diagonal matrix.

[0068] In the above implementation process, by performing singular value decomposition on the preprocessed data matrix, high-dimensional data can be reduced to low-dimensional data, effectively reducing the dimension of the data and reducing the computational complexity. In addition, by retaining the features corresponding to the singular values, important information in the preprocessed data matrix can be retained, improving the accuracy of data interpolation.

[0069] In one possible implementation, the method for obtaining the first diagonal matrix includes: using the PX-MDC model to classify the missing data to obtain the missing type; when the missing type of the missing data is a non-random missing type, randomly weighting the preprocessed data of the non-random missing type to obtain a randomly weighted preprocessed data matrix; performing singular value decomposition on the randomly weighted preprocessed data matrix to obtain a first diagonal matrix containing singular values.

[0070] The PX-MDC model here is a missing data classification model based on particle swarm optimization and XGBoost.

[0071] The above-mentioned non-random missing type means that the probability of missing is independent of any observed or unobserved value, and the non-random missing type is completely random.

[0072] Among them, random weighting refers to selecting random numbers in a set range for weighting. The use of random weighting introduces a certain degree of randomness into singular value decomposition interpolation, which helps to reduce the impact of systematic deviations of non-random missing types and makes the interpolation model more flexible and robust. Among them, the setting range can be set according to actual needs. For example, the setting range can be [0.25, 0.5], the setting range can also be [0.5, 0.75], etc. The setting range can be adjusted according to actual needs.

[0073] In one embodiment, random weighting can be achieved by the following formula: ; in, is the preprocessed data matrix, is the matrix after normal distribution processing, is a random array. The size and number of are consistent with the size and number of non-random missing types.

[0074] The above If the value of is missing non-randomly, The value of is a random number. If the value of is not missing non-randomly, The value of is 1.

[0075] In the above implementation process, when the missing type is non-random missing, by weighting the data before interpolation to give higher weight to the observed data, reliable data can be used to the greatest extent, while the weight of the non-random missing type data can be reduced, reducing the negative impact of missing data on matrix decomposition and improving the flexibility and robustness of the model.

[0076] In one possible implementation, the singular values ​​in the first diagonal matrix are processed based on a set threshold to obtain a second diagonal matrix, including: subtracting the set threshold from all singular values ​​in the first diagonal matrix to obtain an intermediate matrix; retaining a first preset proportion of data in the intermediate matrix to obtain a second diagonal matrix.

[0077] The first preset ratio is the percentage of the sum of the retained singular values ​​to the total singular values.

[0078] It can be understood that after determining the first diagonal matrix, the singular values ​​can be processed by setting a threshold, and then the second diagonal matrix can be obtained by retaining data in which the sum of the singular values ​​accounts for a first preset proportion of the total singular values, thereby achieving the purpose of optimizing the first diagonal matrix.

[0079] The above intermediate matrix can be obtained in the following manner: subtract a set threshold from a singular value, set the singular values ​​less than the set threshold to 0, and obtain an intermediate matrix. The intermediate matrix is ​​also a diagonal matrix.

[0080] The first preset ratio here may be a ratio value set in advance. For example, the first preset ratio may be 50%, 80%, 98%, etc. The first preset ratio may be adjusted according to actual conditions.

[0081] In one embodiment, the threshold is set to 98%. Since the data contains a large amount of information, in order to ensure the rationality of the interpolation result, it is recommended to retain as many singular values ​​as possible, by setting the first preset ratio to 98%. By retaining 98% of the singular values, the first diagonal matrix is ​​optimized to obtain the second diagonal matrix.

[0082] In the above implementation process, after determining the first diagonal matrix, by setting the first preset ratio, the data of the first preset ratio in the intermediate matrix is ​​retained, and by adjusting the value of the first preset ratio, as many singular values ​​as possible can be retained to optimize the interpolation effect.

[0083] In a possible implementation, step 201 includes: replacing missing values ​​with the minimum value of the metabolites in the original data that are not missing; calculating the mean and variance of the metabolites after replacing the missing values; generating normally distributed data based on the mean and variance; performing corresponding preprocessing on the normally distributed data according to the missing type of the missing data to obtain a preprocessed data matrix.

[0084] The metabolite minimum here refers to the minimum value in the current metabolomics data.

[0085] It can be understood that for the missing data in the missing data matrix, the minimum value of the metabolite in the original data that is not missing can be used for filling, and the minimum value of the metabolite in the original data that is not missing can be filled in the missing data position in the missing data matrix.

[0086] The above mean can be calculated as follows: ; in, is the mean, is the number of data in the missing data matrix of the metabolites after replacing missing values, For the data.

[0087] The variance can be calculated as follows: ; in, is the variance, is the number of data in the missing data matrix of the metabolites after replacing missing values, For the data, is the mean.

[0088] The above-mentioned normal distribution data can be generated by setting a function, such as np.random.normal function, randn function, NORM.INV function, etc. The setting function can be selected according to the actual situation.

[0089] It should be understood that missing data of different missing types may have different corresponding processing methods due to different missing reasons. Therefore, using corresponding preprocessing methods for missing data of different missing types can improve the accuracy of missing data processing.

[0090] In the above implementation process, by performing normal distribution processing on the missing data matrix, the data distribution characteristics in the missing data matrix are made more intuitive and easy to explain, the difficulty of processing the missing data matrix is ​​simplified, and the convenience of processing the missing data matrix is ​​improved.

[0091] In a possible implementation, the normally distributed data is preprocessed accordingly according to the missing type of the missing data to obtain a preprocessed data matrix, including: when the missing data is a randomly missing type, the missing data is filled by randomly selecting values ​​from the normally distributed data; when the missing data is a non-random missing type, the missing data is filled by selecting data less than the minimum value of the metabolite from the normally distributed data; and the preprocessed data matrix is ​​obtained through the filled missing data matrix.

[0092] It can be understood that a missing data matrix may include multiple missing data, and the multiple missing data may correspond to one or more missing types. For missing data of different missing types, corresponding filling methods can be used to fill the missing data in the missing data matrix.

[0093] In the above implementation process, when preprocessing the missing data matrix, the missing data is filled in using a corresponding filling method according to the missing type of the missing data, which can improve the filling accuracy.

[0094] In one possible implementation, step 204 includes: determining the missing position of the missing data according to the missing data matrix; determining the missing type of the missing data according to the preprocessed data matrix; determining the missing position and missing type of the missing data in the first approximate matrix according to the missing position and missing type of the missing data; interpolating the corresponding missing positions in the first approximate matrix using a corresponding interpolation method according to the missing type of the missing data in the first approximate matrix to determine the second approximate matrix.

[0095] It can be understood that when interpolating the data in the first approximate matrix, the missing position of the missing data in the missing data matrix can be determined according to the missing data matrix, and then the missing position is matched to the corresponding position in the first approximate matrix. Then, the missing type of the missing data processed by the missing position is determined according to the preprocessed data matrix, and the corresponding interpolation method is determined according to the missing type, and the interpolation data obtained by the interpolation method is interpolated to the corresponding position of the first approximate matrix to complete the data interpolation at the corresponding position of the first approximate matrix. Repeat the above steps until all the missing positions in the missing data matrix are processed in the corresponding position of the first approximate matrix for the corresponding data interpolation, and the interpolation of the first approximate matrix is ​​completed to obtain the second approximate matrix.

[0096] In the above implementation process, when determining the second approximate matrix, the missing position of the missing data is determined according to the missing data matrix, and the missing type of the missing data is determined according to the preprocessed data matrix, and then the corresponding interpolation method is determined to interpolate at the corresponding processing position of the first approximate matrix. The interpolation data can be accurately interpolated to the corresponding position, thereby improving the accuracy of the second approximate matrix.

[0097] In one possible implementation, the corresponding missing positions in the first approximate matrix are interpolated using a corresponding interpolation method according to the missing type of the missing data in the first approximate matrix to determine the second approximate matrix, including: when the missing data is a random missing type, the missing data is filled by randomly selecting values ​​from the normally distributed data.

[0098] When the missing data is of non-random missing type and the value of the missing data position at the corresponding position in the first approximate matrix of the missing data is higher than the minimum value of the metabolite that is not missing in the original data, the missing data is filled by the first weight value of the corresponding position in the preprocessed data matrix and the second weight value of the corresponding position in the first approximate matrix.

[0099] When the missing data are of non-random missing type and the value of the corresponding position of the missing data in the first approximate matrix is ​​not higher than the minimum value of the metabolite in the original data that is not missing, the missing data are filled by the value of the corresponding position in the first approximate matrix.

[0100] In one embodiment, when the missing data is of non-random missing type, the missing data can be interpolated in the following manner: ; in, is the second approximate matrix, is the first approximate matrix, Indicates that the missing value position is not the position where the value of the corresponding position in the first approximation matrix is ​​greater than the minimum value of the metabolite that is not missing in the original data, Indicates that the missing value position is the position where the value of the corresponding position in the first approximation matrix is ​​greater than the minimum value of the metabolite that is not missing in the original data, is the first weight value, is the second weight value.

[0101] The data interpolation method involved in the embodiments of the present application can improve the defects of traditional data interpolation methods in interpolating non-random missing types.

[0102] In order to verify that the data interpolation method in the embodiment of the present application improves the defects of the traditional data interpolation method in the interpolation of non-random missing types, the experiment designed experimental data with a non-random missing type missing data ratio of about 50% to 100% and a step length of about 5%. For each experiment with a non-random missing type ratio, the missing rate ranged from 5% to 40%, with a step length of about 2.5%, and six groups of different concentrations were randomly designed for experiments.

[0103] In order to better evaluate the application of the data interpolation method in the embodiments of the present application in the missing data of metabolomics, real metabolomics data sets are used in the embodiments of the present application, and all data sets can be accessed through the metabolomics workbench (https: / / www.metabolomicsworkbench.org / ).

[0104] Specifically, this dataset comes from two different experiments, one of which includes two datasets, namely asthma / obesity related research (mouse) and protein family data (bacteria). Among them, the asthma / obesity related research (mouse) dataset is used to study how three different diets affect the arginine metabolic pathway in the lungs of mice. The experiment contained 80 mouse lung tissue samples, which had 612 metabolites measured by GC-MS. The dataset ID is ST000419. Protein family data (bacteria) is used to examine the differences in intracellular metabolite concentrations between wild-type and knockout strains. This experiment included 5 groups, each with 6 replicates, of which 1 sample was missing (29 samples in total), and 249 metabolites were measured by GC-MS. The dataset ID is ST000118.

[0105] In order to quantify the interpolation error, the normalized root mean square error is often used to evaluate the difference between the interpolated value and the true value. The calculation formula is as follows: ; in, represents the value of the real data, and is the data value after interpolation. NRMSE can measure the standardized level of error between the interpolated value and the true value and is sensitive to outliers.

[0106] Considering that the missing values ​​of non-random missing types are not randomly distributed, directly using NRMSE evaluation may lead to biased results. The traditional approach is to first calculate the NRMSE of each missing variable, rank them in different interpolation methods, sum the ranks of all missing variables of each method, and compare them based on SOR. SOR can be expressed by the following formula: ; in, is the number of missing variables, Refers to different interpolation methods in of the missing variables rank.

[0107] like Figure 3-Figure 6 As shown, in this study, random weighting is to select random numbers in [0.5, 0.75] for weighting. By comparing the data interpolation method in the embodiment of the present application ( Figure 3-Figure 6CWSVD_13) and traditional data interpolation methods [including: SVD ( Figure 3-Figure 6 soft_svd), random forest ( Figure 3-Figure 6 rf), K nearest neighbors ( Figure 3-Figure 6 ns_knn) and QRILC] shown in [ ] under different missing rates and non-random missing type ratios, it is found that the data interpolation method in the embodiment of the present application is significantly better than other methods in both NRMSE and SOR evaluation indicators. Especially in the case of high missing rates (over 20%) and high non-random missing type ratios (over 50%), the NRMSE (Chinese name: Normalized Root Mean Square Error) value and SOR (Chinese name: Rank Sum) score of the data interpolation method in the embodiment of the present application are much lower than other methods, showing more outstanding robustness and data recovery accuracy. Under extreme test conditions where the non-random missing type ratio is close to 99%, the performance of the data interpolation method in the embodiment of the present application is stable, while the errors of other methods increase sharply.

[0108] The above comparison not only proves the efficiency and reliability of the data interpolation method in the embodiment of the present application in processing complex non-random missing data patterns, but also highlights its advantages in accurately interpolating missing data, which is crucial for data integrity and accuracy of analysis.

[0109] In the above implementation process, for missing data of different missing types, corresponding interpolation methods are used to interpolate missing data, which can improve the accuracy of missing data interpolation, increase the missing types that can be targeted by the data interpolation method, and increase the application scenarios.

[0110] Based on the same application concept, a metabolic state determination device corresponding to the metabolic state determination method is also provided in the embodiment of the present application. Since the principle of solving the problem by the device in the embodiment of the present application is similar to that in the aforementioned metabolic state determination method embodiment, the implementation of the device in this embodiment can refer to the description in the embodiment of the above method, and the repeated parts will not be repeated.

[0111] See also Figure 7 , is a functional module diagram of the metabolic state determination device provided in the embodiment of the present application. Each module in the metabolic state determination device in this embodiment is used to execute each step in the above method embodiment. The metabolic state determination device includes a preprocessing module 301, a decomposition module 302, an update module 303, an interpolation module 304, an iteration module 305, and a determination module 306; wherein, The preprocessing module 301 is used to preprocess the missing data matrix of metabolomics using normal distribution to obtain a preprocessed data matrix; wherein the metabolomics is configured to reveal the metabolic state of an organism under different physiological or pathological states.

[0112] The decomposition module 302 is used to perform singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix.

[0113] The updating module 303 is used to update the missing data matrix according to the diagonal matrix to obtain a first approximate matrix.

[0114] The interpolation module 304 is used to interpolate the data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix.

[0115] The iteration module 305 is used to repeat the process of performing singular value decomposition using the second approximate matrix as the preprocessed data matrix until an iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result.

[0116] The determination module 306 is used to determine the metabolic state of the organism according to the interpolated metabolomics data.

[0117] In a possible implementation, the decomposition module 302 is further used to: perform singular value decomposition on the preprocessed data matrix to obtain a first diagonal matrix containing singular values; and process the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix.

[0118] In a possible implementation manner, the updating module 303 is specifically configured to: update the missing data matrix according to the second diagonal matrix to obtain a first approximate matrix.

[0119] In one possible implementation, the decomposition module 302 is specifically used to: classify the missing data using the PX-MDC model to obtain the missing type; when the missing type of the missing data is a non-random missing type, randomly weight the preprocessed data of the non-random missing type to obtain a randomly weighted preprocessed data matrix; perform singular value decomposition on the randomly weighted preprocessed data matrix to obtain a first diagonal matrix containing singular values.

[0120] In one possible implementation, the decomposition module 302 is specifically used to: obtain an intermediate matrix by subtracting the set threshold from all singular values ​​in the first diagonal matrix; retain a first preset proportion of data in the intermediate matrix to obtain the second diagonal matrix; wherein the first preset proportion is the percentage of the sum of the retained singular values ​​to the total singular values.

[0121] In one possible implementation, the preprocessing module 301 is further used to: replace missing values ​​with the minimum value of the metabolites that are not missing in the original data; calculate the mean and variance of the metabolites after replacing the missing values; generate normally distributed data based on the mean and the variance; and perform corresponding preprocessing on the normally distributed data according to the missing type of the missing data to obtain a preprocessed data matrix.

[0122] In one possible implementation, the preprocessing module 301 is specifically used to: when the missing data is a randomly missing type, fill the missing data by randomly selecting values ​​from the normally distributed data; when the missing data is a non-randomly missing type, select data less than the minimum value of the metabolite from the normally distributed data to fill the missing data; and obtain the preprocessing data matrix through the missing data matrix after filling.

[0123] In one possible implementation, the interpolation module 304 is specifically used to: determine the missing position of the missing data according to the missing data matrix; determine the missing type of the missing data according to the preprocessed data matrix; determine the missing position and missing type of the missing data in the first approximate matrix through the missing position and missing type of the missing data; interpolate the corresponding missing positions in the first approximate matrix using a corresponding interpolation method according to the missing type of the missing data in the first approximate matrix to determine the second approximate matrix.

[0124] In one possible implementation, the interpolation module 304 is specifically used to: when the missing data is a randomly missing type, fill the missing data by using a randomly selected value in the normally distributed data; when the missing data is a non-random missing type, and the value of the missing data position of the missing data at the corresponding position in the first approximate matrix is ​​higher than the minimum value of the metabolite that is not missing in the original data, fill the missing data by using a first weight value at the corresponding position of the preprocessed data matrix and a second weight value at the corresponding position in the first approximate matrix; when the missing data is a non-random missing type, and the value of the missing data at the corresponding position in the first approximate matrix is ​​not higher than the minimum value of the metabolite that is not missing in the original data, fill the missing data by using the value of the corresponding position in the first approximate matrix.

[0125] In addition, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the metabolic state determination method described in the above method embodiment are executed.

[0126] The computer program product of the metabolic state determination method provided in the embodiments of the present application includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the steps of the metabolic state determination method described in the above method embodiments. For details, please refer to the above method embodiments, which will not be repeated here.

[0127] In several embodiments provided in the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely schematic. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0128] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0129] If the function is implemented in the form of a software function module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art, and the computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes. It should be noted that in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises" or any other variation thereof are intended to cover non-exclusive inclusion, so that a process, method, article or apparatus that includes a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or apparatus. In the absence of more restrictions, the elements defined by the sentence "includes..." do not exclude the presence of other identical elements in the process, method, article or apparatus that includes the elements. The above description is only a preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application. It should be noted that similar numbers and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.

[0130] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for determining a metabolic state, characterized in that: include: Preprocessing a missing data matrix of metabolomics using normal distribution to obtain a preprocessed data matrix; wherein the metabolomics is configured to reveal the metabolic state of an organism under different physiological or pathological states; Performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix; Update the missing data matrix according to the diagonal matrix to obtain a first approximate matrix; interpolating the data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix; Repeating from taking the second approximate matrix as the preprocessed data matrix to perform singular value decomposition until an iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result; Determine the metabolic status of an organism based on the imputed metabolomics data.

2. The method according to claim 1, characterized in that The diagonal matrix includes a first diagonal matrix and a second diagonal matrix; The step of performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix includes: Performing singular value decomposition on the preprocessed data matrix to obtain a first diagonal matrix containing singular values; Processing the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix; The step of updating the missing data matrix according to the diagonal matrix to obtain a first approximate matrix comprises: The missing data matrix is ​​updated according to the second diagonal matrix to obtain a first approximate matrix.

3. The method according to claim 2, characterized in that in, The first diagonal matrix is ​​obtained by: Using the PX-MDC model to classify the missing data, obtaining the missing type; In the case where the missing data is of a non-random missing type, randomly weighting the pre-processed data of the non-random missing type to obtain a pre-processed data matrix after random weighting; Perform singular value decomposition on the randomly weighted preprocessed data matrix to obtain the first diagonal matrix containing singular values.

4. The method according to claim 2, characterized in that: The step of processing the singular values ​​in the first diagonal matrix based on a set threshold to obtain a second diagonal matrix includes: Subtracting the set threshold from all singular values ​​in the first diagonal matrix to obtain an intermediate matrix; Retaining a first preset proportion of data in the intermediate matrix to obtain the second diagonal matrix; The first preset ratio is the percentage of the sum of the retained singular values ​​to the total singular values.

5. The method according to claim 1, characterized in that The method uses normal distribution to preprocess the missing data matrix of metabolomics to obtain a preprocessed data matrix, including: Missing values ​​were replaced by the minimum value of the metabolite that was not missing in the original data; Calculate the mean and variance of metabolites after replacing missing values; generating normally distributed data based on the mean and the variance; The normally distributed data are preprocessed accordingly according to the missing type of the missing data to obtain a preprocessed data matrix.

6. The method according to claim 5, characterized in that The method of performing corresponding preprocessing on the normally distributed data according to the missing type of the missing data to obtain a preprocessed data matrix includes: In the case where the missing data is of a randomly missing type, the missing data is filled by randomly selecting values ​​from the normally distributed data; In the case where the missing data is of non-random missing type, data less than the minimum value of the metabolite is selected from the normally distributed data to fill the missing data; The preprocessed data matrix is ​​obtained by filling in the missing data matrix.

7. The method according to any one of claims 1 to 6, characterized in that: The interpolating the data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix comprises: Determine the missing position of the missing data according to the missing data matrix; Determining the missing type of missing data according to the preprocessed data matrix; Determine the missing position and missing type of the missing data in the first approximate matrix according to the missing position and the missing type of the missing data; The corresponding missing positions in the first approximate matrix are interpolated using a corresponding interpolation method according to the missing type of the missing data in the first approximate matrix to determine the second approximate matrix.

8. The method according to claim 7, characterized in that The step of interpolating corresponding missing positions in the first approximate matrix using corresponding interpolation methods according to missing types of missing data in the first approximate matrix to determine the second approximate matrix includes: In the case where the missing data is of the randomly missing type, the missing data is filled by randomly selecting values ​​from the normally distributed data; When the missing data is of a non-random missing type and the value of the missing data position of the missing data at the corresponding position in the first approximate matrix is ​​higher than the minimum value of the metabolite that is not missing in the original data, the missing data is filled by using the first weight value of the corresponding position in the preprocessed data matrix and the second weight value of the corresponding position in the first approximate matrix; When the missing data is of non-random missing type and the value of the corresponding position of the missing data in the first approximate matrix is ​​not higher than the minimum value of the metabolite in the original data that is not missing, the missing data is filled by the value of the corresponding position in the first approximate matrix.

9. A metabolic state determination device, characterized in that: include: A preprocessing module, for preprocessing a missing data matrix of metabolomics using a normal distribution to obtain a preprocessed data matrix; wherein the metabolomics is configured to reveal the metabolic state of an organism under different physiological or pathological states; A decomposition module, used for performing singular value decomposition on the preprocessed data matrix to obtain a diagonal matrix; An updating module, used for updating the missing data matrix according to the diagonal matrix to obtain a first approximate matrix; an interpolation module, configured to interpolate data in the first approximate matrix according to the missing data matrix, the preprocessed data matrix and the first approximate matrix to obtain a second approximate matrix; An iteration module, used to repeat from taking the second approximate matrix as the preprocessed data matrix to perform singular value decomposition until an iteration condition is met, and the second approximate matrix finally obtained is the final interpolation result; The determination module is used to determine the metabolic state of an organism based on the imputed metabolomics data.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are executed.

Images