Method, device and medium for determining a chromatography prediction model and predicting chromatography data

Abstract

The application discloses a chromatography prediction model determination method and device, electronic equipment and storage medium. The method comprises the following steps: obtaining a chromatography mechanism model under a set chromatography scene; determining a target influencing factor among a plurality of undetermined influencing factors under the set chromatography scene based on the chromatography mechanism model; obtaining first sample data obtained by the chromatography mechanism model for the target influencing factor; and obtaining second sample data obtained by a wet experiment for the target influencing factor; the first sample data and the second sample data respectively comprise a data value of the target influencing factor and a data value of a chromatography target item; a chromatography prediction model is constructed based on the first sample data and the second sample data, and the chromatography prediction model represents a mapping relationship between the chromatography target item and the target influencing factor. The application realizes the effect of constructing a chromatography prediction model with higher chromatography prediction accuracy by combining chromatography prediction data of the chromatography mechanism model and real experimental data of the wet experiment.

Description

Technical Field

[0001] This invention relates to the field of computer application technology, and in particular to a method, apparatus and medium for determining a tomographic prediction model and predicting tomographic data. Background Technology

[0002] Currently, chromatography, with its high separation efficiency, is widely used in substance separation and purification, component analysis, quality control, and pollutant monitoring. In practical applications, it is usually necessary to determine the chromatographic conditions that achieve the desired chromatographic effect based on the predictive capabilities of a predictive model before performing chromatographic separation.

[0003] In existing technologies, prediction models are usually constructed using sample data formed by tomographic mechanism models. However, current tomographic mechanism models often use simple methods such as averaging to characterize the separation process, which often introduces errors into the prediction model, resulting in differences between the prediction results of the prediction model and the actual wet experimental results. Summary of the Invention

[0004] This invention provides a method, apparatus, and medium for determining a tomographic prediction model and predicting tomographic data, in order to solve the technical problem that the tomographic prediction accuracy of traditional prediction models based on tomographic mechanism models is poor.

[0005] According to one aspect of the present invention, a method for determining a tomographic prediction model is provided, the method comprising:

[0006] Obtain the tomography mechanism model under the specified tomography scenario;

[0007] Based on the tomography mechanism model, the target influencing factor is determined from multiple undetermined influencing factors in a given tomography scenario;

[0008] The process involves obtaining first sample data for the target influencing factor through the chromatographic mechanism model; and obtaining second sample data for the target influencing factor through a wet experiment; the first sample data and the second sample data respectively include data values ​​of the target influencing factor and data values ​​of the chromatographic target item.

[0009] A chromatographic prediction model is constructed based on the first sample data and the second sample data. The chromatographic prediction model represents the mapping relationship between the chromatographic target item and the target influencing factors.

[0010] According to another aspect of the present invention, a tomographic prediction model determination apparatus is provided, the apparatus comprising:

[0011] The mechanism model determination module is used to obtain the tomography mechanism model under the given tomography scenario;

[0012] The influencing factor screening module is used to determine the target influencing factor from multiple undetermined influencing factors under a set chromatographic scenario based on the chromatographic mechanism model.

[0013] The sample data determination module is used to acquire first sample data obtained through the chromatographic mechanism model for the target influencing factor; and to acquire second sample data obtained through wet experiment for the target influencing factor; the first sample data and the second sample data respectively include the data value of the target influencing factor and the data value of the chromatographic target item;

[0014] The prediction model building module is used to build a stratified prediction model based on the first sample data and the second sample data. The stratified prediction model represents the mapping relationship between the stratified target item and the target influencing factors.

[0015] According to another aspect of the present invention, a method for predicting tomographic data is provided, the method comprising:

[0016] Determine the target input values ​​for the target influencing factors in the tomographic scenario;

[0017] Determine the target predicted value of the chromatographic target item based on the target input value of the target influencing factor determined by the chromatographic prediction model.

[0018] According to one aspect of the present invention, a tomographic data prediction apparatus is provided, the apparatus comprising:

[0019] The input value determination module is used to determine the target input values ​​of the target influencing factors in the set tomographic scenario;

[0020] The stratified prediction module is used to determine the target predicted value of the stratified target item determined by the stratified prediction model for the target influencing factors based on the target input value.

[0021] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0022] At least one processor; and

[0023] A memory communicatively connected to the at least one processor; wherein,

[0024] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method for determining the tomographic prediction model or the tomographic data prediction method according to any embodiment of the present invention.

[0025] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions, the computer instructions being configured to cause a processor to execute and implement the method for determining a tomographic prediction model or the tomographic data prediction method according to any embodiment of the present invention.

[0026] The technical solution of this invention involves: acquiring a tomographic mechanism model under a given tomographic scenario; determining a target influencing factor from multiple undetermined influencing factors under the given tomographic scenario based on the tomographic mechanism model; acquiring first sample data obtained through the tomographic mechanism model for the target influencing factor; and acquiring second sample data obtained through wet experiments for the target influencing factor. The first and second sample data respectively include data values ​​of the target influencing factor and data values ​​of the tomographic target item. A tomographic prediction model is constructed based on the first and second sample data, and the tomographic prediction model characterizes the mapping relationship between the tomographic target item and the target influencing factor. This solves the technical problem that the tomographic mechanism model is limited by computational solving capabilities, resulting in poor tomographic prediction accuracy. This invention achieves the effect of constructing a tomographic prediction model with higher accuracy by combining the tomographic prediction data of the tomographic mechanism model and the real experimental data of wet experiments.

[0027] The technical solution of this invention involves determining the target input value of a target influencing factor under a defined tomographic scenario; and determining the target predicted value of the tomographic target item based on the target input value of the target influencing factor using a tomographic prediction model. This invention improves the accuracy of tomographic effect prediction under tomographic separation conditions using the constructed tomographic prediction model.

[0028] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a flowchart of a method for determining a tomographic prediction model according to Embodiment 1 of the present invention;

[0031] Figure 2 This is a flowchart of a method for determining a tomographic prediction model according to Embodiment 2 of the present invention;

[0032] Figure 3 This is an overall flowchart of a method for determining a tomographic prediction model according to an embodiment of the present invention;

[0033] Figure 4 This is a prediction effect diagram of a preliminary prediction model provided according to an embodiment of the present invention.

[0034] Figure 5 This is a prediction effect diagram of a tomographic prediction model provided according to an embodiment of the present invention.

[0035] Figure 6 This is a schematic diagram of the structure of a device for determining a tomographic prediction model according to Embodiment 3 of the present invention;

[0036] Figure 7 This is a flowchart of a tomographic data prediction method provided in Embodiment 4 of the present invention;

[0037] Figure 8 This is a schematic diagram of the structure of a tomographic data prediction device provided in Embodiment 5 of the present invention;

[0038] Figure 9 This is a schematic diagram of the structure of an electronic device that implements the method for determining the tomographic prediction model or the tomographic data prediction method of the embodiments of the present invention. Detailed Implementation

[0039] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0040] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0041] Example 1

[0042] Figure 1 This is a flowchart illustrating a method for determining a tomographic prediction model according to Embodiment 1 of the present invention. This embodiment is applicable to situations where the predictive power of a tomographic mechanism model is optimized to construct a tomographic prediction model with higher predictive power. This method can be executed by a tomographic prediction model determining device, which can be implemented in hardware and / or software and can be configured in a computer. Figure 1 As shown, the method includes:

[0043] S110. Obtain the tomography mechanism model under the set tomography scenario.

[0044] The chromatography mechanism model can be understood as a model capable of performing chromatography mechanism analysis. This chromatography mechanism model can be a system of partial differential equations comprising physicochemical equations such as molecular diffusion equations, convection-diffusion equations, and adsorption kinetics equations. In this embodiment of the invention, the chromatography mechanism model can be pre-constructed for a specific chromatography scenario.

[0045] The defined chromatography scenario can be understood as a scenario for performing chromatographic separation on a defined sample. In this embodiment of the invention, one defined chromatography scenario can correspond to one defined sample. The defined sample can be preset according to scenario requirements and is not specifically limited here. Optionally, the defined sample can be a protein, an organic compound, or airborne particulate matter, etc. For example, a protein can be a monoclonal antibody, an antibody fragment, a recombinant protein, or a plasma protein, etc. An organic compound can be a chiral molecule or a synthetic intermediate, etc. Airborne particulate matter can be a polycyclic aromatic hydrocarbon or a sulfate, etc.

[0046] S120. Based on the tomography mechanism model, determine the target influencing factor from among multiple undetermined influencing factors in the set tomography scenario.

[0047] The "undetermined influencing factors" can be understood as factors whose degree of influence on the set chromatography scenario needs to be determined. In this embodiment of the invention, the undetermined influencing factors can be all factors related to the set chromatography scenario, or factors obtained after manually screening all factors. For example, taking a monoclonal antibody as the set sample corresponding to the set chromatography scenario, the corresponding undetermined influencing factors may include sample loading capacity, sample flow rate, elution pH, elution salt concentration, and packing pore size, etc.

[0048] The target influencing factor can be understood as a factor that has a strong impact on the chromatographic separation of the set chromatography scenario. For example, taking a monoclonal antibody as the set sample corresponding to the set chromatography scenario, the target influencing factor may be the loading capacity, elution pH, and elution salt concentration among the factors to be determined.

[0049] Based on the above embodiments, optionally, determining the target influencing factor from multiple undetermined influencing factors in a given tomographic scenario based on the tomographic mechanism model includes:

[0050] Based on the tomography mechanism model, determine the degree of influence of each of the undetermined influencing factors on the tomography target item under the given tomography scenario;

[0051] The target influencing factor is determined from among the multiple undetermined influencing factors based on the degree of influence.

[0052] In this embodiment of the invention, the degree of influence of different undetermined influencing factors on the chromatographic target item can be different. Optionally, the number of target influencing factors can be preset according to scenario requirements, or it can be determined according to a threshold related to the degree of influence. Specifically, the undetermined influencing factors can be sorted in descending order based on the degree of influence, and the top three undetermined influencing factors in the sorted result can be used as target influencing factors. Alternatively, at least one undetermined influencing factor whose degree of influence exceeds the threshold can be used as a target influencing factor.

[0053] Optionally, based on the above embodiments, determining the degree of influence of each undetermined influencing factor on the chromatographic target item under the set chromatographic scenario based on the chromatographic mechanism model includes: determining multiple first chromatographic conditions corresponding to multiple undetermined influencing factors through statistical experimental design methods, each first chromatographic condition including data values ​​corresponding to multiple undetermined influencing factors respectively; determining the first target value of the chromatographic target item under each first chromatographic condition based on the chromatographic mechanism model, and determining the degree of influence of each undetermined influencing factor on the chromatographic target item according to the first target value of the chromatographic target item corresponding to different chromatographic conditions respectively.

[0054] The statistical design of experiment (DoE) method can quantify the interaction between multiple undetermined influencing factors and output multiple stratification conditions. It should be noted that, for example, if there are 5 undetermined influencing factors, and each undetermined influencing factor corresponds to three data values, then without considering the interaction between factors, there are 125 possible stratification conditions. However, by using the statistical design of experiment method to determine the synergistic effect between the undetermined influencing factors for each stratification condition, the number of possible first stratification conditions can be less than 125. In this embodiment of the invention, the statistical design of experiment method can be preset according to the scenario requirements, and is not specifically limited here. Optionally, the statistical design of experiment method may include, but is not limited to, factorial design (ffd), response surface methodology (RSM), screening design, and custom design.

[0055] The first chromatography condition can be understood as the chromatography separation condition constructed for the unknown influencing factors. The chromatography target term can be understood as the output data term of the chromatography mechanism model. The first target value can be understood as the specific value corresponding to the output data term. For example, the chromatography target term can be the sample yield. The first target value can be the yield value.

[0056] Specifically, determining the degree of influence of each undetermined influencing factor on the chromatographic target item based on the first target value corresponding to different chromatographic conditions may include:

[0057] For any undetermined influencing factor, multiple first-level conditions are determined, including different data values ​​of the undetermined influencing factor, and a first target value corresponding to each of the multiple first-level conditions. The undetermined influencing factor is a unique variable among the multiple first-level conditions. The degree of influence of the undetermined influencing factor on the chromatographic target item is determined based on the minimum and maximum values ​​of the first target values. Specifically, the degree of influence of the undetermined influencing factor on the chromatographic target item can be determined based on the difference between the minimum and maximum values ​​of the first target values; the larger the difference, the greater the degree of influence of the undetermined influencing factor on the chromatographic target item.

[0058] For example, the differences corresponding to each undetermined influencing factor can be sorted, and the top n undetermined influencing factors can be selected as target influencing factors, or the undetermined influencing factors with differences greater than the difference threshold can be determined as target influencing factors.

[0059] Based on the above embodiments, the number of first tomographic conditions can be effectively reduced, thereby further reducing the computational load of tomographic mechanism analysis of each first tomographic condition based on the tomographic mechanism model and improving the efficiency of determining the degree of influence of each undetermined influencing factor on the tomographic target item.

[0060] Based on the above embodiments, optionally, determining the degree of influence of each of the undetermined influencing factors on the tomographic target item under the set tomographic scenario based on the tomographic mechanism model includes:

[0061] The first chromatographic condition is determined by iterating through the data values ​​corresponding to multiple undetermined influencing factors. Based on the chromatographic mechanism model, the first target value of the chromatographic target item is determined under each of the first chromatographic conditions. The degree of influence of each undetermined influencing factor on the chromatographic target item is then determined according to the first target value of the chromatographic target item corresponding to different chromatographic conditions. The method for determining the degree of influence of the undetermined influencing factors on the chromatographic target item is not elaborated here.

[0062] In this embodiment, by screening target influencing factors from the undetermined influencing factors, interference items in the tomographic scenario are removed, thereby improving the accuracy of influencing factors in the tomographic prediction model.

[0063] S130. Obtain first sample data for the target influencing factor obtained through the chromatographic mechanism model; and obtain second sample data for the target influencing factor obtained through wet experiment; the first sample data and the second sample data respectively include the data value of the target influencing factor and the data value of the chromatographic target item.

[0064] Here, the first sample data can be understood as sample data determined through dry experiments. Dry experiments involve processing the data values ​​of the target influencing factors using a chromatographic mechanism model to achieve a chromatographic simulation data processing process. Dry experiments are fast and can quickly obtain a large amount of first sample data.

[0065] The wet test can be understood as an actual experiment conducted targeting the influencing factors, that is, an actual chromatographic experiment performed by adjusting the chromatographic equipment, sample, and chromatographic solution to control the influencing factors. The wet test yields real sample data, improving the authenticity and accuracy of the data.

[0066] The second sample data can be understood as sample data determined through a wet experiment. In this embodiment of the invention, the data value determined through the wet experiment is taken as the true value. That is, the second sample data is taken as the true sample data.

[0067] Optionally, obtaining the first sample data for the target influencing factor through the stratification mechanism model includes: determining a plurality of second stratification conditions corresponding to the target influencing factor, each second stratification condition including the data value of the target influencing factor; obtaining the second target value of the stratification target item for each second stratification condition through the stratification mechanism model; and forming the first sample data based on each second stratification condition and the second target value of the stratification target item.

[0068] The second chromatographic condition can be understood as a chromatographic separation condition constructed for the target influencing factors. In this invention, the number of second chromatographic conditions is less than the number of first chromatographic conditions. The specific values ​​of the second target value and the first target value are of the same category, which will not be elaborated here. Specifically, forming the first sample data based on each second chromatographic condition and the second target value of the chromatographic target item includes: for each second chromatographic condition, generating a first subsample based on the second chromatographic condition and the second target value of the chromatographic target item corresponding to the second chromatographic condition; and forming the first sample data based on multiple first subsamples.

[0069] Based on the above embodiments, the rapid construction of the first sample data using existing tomographic mechanism models is achieved.

[0070] Optionally, obtaining the second sample data obtained through wet experiment for the target influencing factor includes: classifying the second chromatographic conditions based on the second target value to obtain multiple data classes; determining a third chromatographic condition in the second chromatographic conditions of each data class; obtaining a third target value of the chromatographic target item obtained through wet experiment for each of the third chromatographic conditions; and forming the second sample data according to each of the third chromatographic conditions and the third target value of the chromatographic target item.

[0071] The data class can characterize the chromatographic separation effect corresponding to the chromatographic conditions. Optionally, the data class can include a Level A separation class, a Level B separation class, and a Level C separation class. In simpler terms, a Level A separation class indicates that the second chromatographic conditions can achieve a high level of chromatographic separation. A Level B separation class indicates that the second chromatographic conditions can achieve a medium level of chromatographic separation. A Level C separation class indicates that the second chromatographic conditions can achieve a low level of chromatographic separation.

[0072] Optionally, determining the third layering condition from the second layering conditions in each data class includes: extracting a target number of second layering conditions as third layering conditions in each data class. The target number can be preset according to scenario requirements and is not specifically limited here. Preferably, the target number can be 1 to 3. For example, if the target number is 2, then two layering conditions are randomly extracted from the second layering conditions of the A-level separation class, two from the second layering conditions of the B-level separation class, and two from the second layering conditions of the C-level separation class, resulting in 6 third layering conditions. In this embodiment of the invention, the number of third layering conditions can be much less than the number of second layering conditions. The number of third layering conditions determined in different data classes can be the same or different. By determining the third layering conditions in different data classes, the comprehensiveness of the distribution of the third layering conditions in different data classes can be improved, and the number of third layering conditions can be reduced, thereby reducing the time consumption of wet testing and improving processing efficiency.

[0073] Specifically, the step of forming the second sample data based on each of the third chromatographic conditions and the third target value of the chromatographic target item includes: generating a second subsample for each third chromatographic condition based on the third chromatographic condition and the third target value of the chromatographic target item corresponding to the third chromatographic condition; and forming the second sample data based on multiple second subsamples.

[0074] Based on the above embodiments, by extracting the third chromatography condition from the second chromatography condition for wet chromatography, the number of chromatography conditions required for wet chromatography can be effectively reduced. Extracting the third chromatography condition through data classification ensures the balance of the extracted third chromatography conditions.

[0075] S140. Construct a stratified prediction model based on the first sample data and the second sample data. The stratified prediction model characterizes the mapping relationship between the stratified target item and the target influencing factor.

[0076] The stratified prediction model is a model with stratified data analysis capabilities. Optionally, the stratified prediction model can be a linear regression model. The stratified prediction model can be obtained by correcting a preliminary prediction model built based on the first sample data using second sample data. The preliminary prediction model can also be a linear regression model. In this embodiment of the invention, compared to the preliminary prediction model, the stratified prediction model includes newly added influencing factors.

[0077] The technical solution of this invention involves: acquiring a tomographic mechanism model under a given tomographic scenario; determining a target influencing factor from multiple undetermined influencing factors under the given tomographic scenario based on the tomographic mechanism model; acquiring first sample data obtained through the tomographic mechanism model for the target influencing factor; and acquiring second sample data obtained through wet experiments for the target influencing factor. The first and second sample data respectively include data values ​​of the target influencing factor and data values ​​of the tomographic target item. A tomographic prediction model is constructed based on the first and second sample data, and the tomographic prediction model characterizes the mapping relationship between the tomographic target item and the target influencing factor. This solves the technical problem that the tomographic mechanism model is limited by computational solving capabilities, resulting in poor tomographic prediction accuracy. This invention achieves the effect of constructing a tomographic prediction model with higher accuracy by combining the tomographic prediction data of the tomographic mechanism model and the real experimental data of wet experiments.

[0078] Example 2

[0079] Figure 2 This is a flowchart illustrating a method for determining a tomographic prediction model according to Embodiment 2 of the present invention. This embodiment refines the tomographic prediction model constructed based on the first sample data and the second sample data described in the above embodiments. Figure 2 As shown, the method includes:

[0080] S210. Obtain the tomography mechanism model under the set tomography scenario.

[0081] S220. Based on the tomography mechanism model, determine the target influencing factor from among multiple undetermined influencing factors in a given tomography scenario.

[0082] S230. Obtain first sample data for the target influencing factor obtained through the chromatographic mechanism model; and obtain second sample data for the target influencing factor obtained through wet experiment; the first sample data and the second sample data respectively include the data value of the target influencing factor and the data value of the chromatographic target item.

[0083] S240. Construct a preliminary prediction model based on the first sample data. The preliminary prediction model represents the preliminary mapping relationship between the chromatographic target item and the target influencing factors.

[0084] Specifically, an impact factor item is constructed based on the target influencing factors; an impact factor item is constructed based on multiple impact factor items; the factor item coefficient corresponding to each impact factor item is determined based on the first sample data; and a preliminary prediction model is constructed based on the impact factor items and factor item coefficients.

[0085] S250. Set data source labels for the first sample data and the second sample data respectively;

[0086] The data source label can characterize the data source of the sample data. Optionally, the data source label includes a first label and a second label. The first label characterizes the data source of the sample data as a chromatographic mechanism model. The second label characterizes the data source of the sample data as a wet experiment.

[0087] S260. The preliminary prediction model is adjusted based on the first sample data and the second sample data carrying the data source label to obtain a stratified prediction model, wherein the stratified prediction model characterizes the mapping relationship between the stratified target item and the data source and the target influencing factors.

[0088] As an optional embodiment, optionally, adjusting the preliminary prediction model based on the first sample data and the second sample data carrying the data source label to obtain the tomographic prediction model includes:

[0089] Based on the preliminary prediction model, the data source is added as a new influencing factor. The preliminary prediction model is adjusted based on the first sample data and the second sample data carrying the data source label to obtain the chromatographic prediction model.

[0090] Based on the above embodiments, the ease of constructing a tomographic prediction model can be improved by directly adjusting the preliminary prediction model based on two types of sample data.

[0091] As an alternative embodiment, optionally, adjusting the preliminary prediction model based on the first sample data and the second sample data carrying the data source label to obtain the stratified prediction model includes: aligning the first sample data and the second sample data carrying the data source label to obtain aligned sample data, wherein the aligned sample data includes the data source label, the data value of the target influencing factor, and the aligned data value of the stratified target item; and, based on the preliminary prediction model, adding the data source as a new influencing factor, adjusting the preliminary prediction model based on the aligned sample data to obtain the stratified prediction model.

[0092] The aligned sample data can be understood as the sample data obtained after data alignment. In this embodiment of the invention, the method of data alignment for the sample data can be preset according to the needs of the scenario, and is not specifically limited here. Optionally, the data alignment method can be linear data alignment, decimal scaling, or logarithmic transformation, etc. The aligned data value can be understood as the data value obtained by aligning the target value of the target influencing factor.

[0093] Specifically, the data values ​​of the same chromatographic conditions corresponding to the target items in the first and second sample data are aligned to obtain aligned sample data. For example, a wet test is performed on the two chromatographic conditions corresponding to the lowest data value (e.g., data value A) and the highest data value (e.g., data value B) of the target item, yielding wet test results (also two data values), which are then corrected accordingly to data values ​​A and B. Furthermore, adjustments can also be made for target items corresponding to chromatographic conditions other than the two mentioned above that fall between A and B.

[0094] Optionally, the wet experiment results can be corrected to data values ​​A and B accordingly. This can be achieved through data alignment or by first constructing a linear fitting relationship and then transforming it, thus adjusting the preliminary prediction model to obtain the tomographic prediction model. Furthermore, the application method of constructing a linear fitting relationship is the same as that of data alignment, and will not be elaborated upon here.

[0095] It should be noted that after obtaining the tomographic prediction model and the target value of the tomographic target item based on the tomographic prediction model, it is necessary to perform reverse data alignment on the target value according to the corresponding coefficients during data alignment in order to update the target value of the above tomographic target item and improve the accuracy of the target value of the tomographic target item.

[0096] The newly added influencing factors can be understood as newly introduced influencing factors. Specifically, an error adjustment term corresponding to the newly added influencing factors is constructed in the preliminary prediction model. The adjustment term coefficient corresponding to the error adjustment term is determined based on the data obtained through wet experiments, thus obtaining the tomographic prediction model.

[0097] Based on the above embodiments, by first aligning the sample data and then adjusting the preliminary prediction model, the prediction accuracy of the constructed tomographic prediction model can be improved.

[0098] Based on the above embodiments, a high-precision tomographic prediction model is constructed by building a preliminary prediction model based on the first sample data automatically generated by the tomographic mechanism model, and then correcting the preliminary prediction model with a small amount of wet experimental data.

[0099] Figure 3 This is an overall flowchart of a method for determining a tomographic prediction model according to an embodiment of the present invention. The following is in conjunction with... Figure 3 The overall process of determining a tomographic prediction model is described.

[0100] Specifically:

[0101] 1. Separate samples under a given chromatography scenario (e.g., high polymer separation commonly seen in biopharmaceuticals) and construct a chromatography mechanism model.

[0102] a) Determine the parameters of the chromatography system and column. Specifically, determine the system piping, dead volume of the mixing tank, column dimensions, interstitial porosity, total porosity, radial diffusion coefficient, axial diffusion coefficient, packing type, and packing size.

[0103] b) Conduct gradient experiments. Specifically, input the chromatographic data into the calculation software to obtain parameters such as Charge and Equilibrium for the adsorption model. For example, the adsorption model can be a size exclusion-mass action (SMA) model. It is important to understand that the parameters can differ between different adsorption models.

[0104] Alternatively, conduct high-load experiments. Collect the elution product in fractions and determine the proportion of different components in each collected product (e.g., main peak, polymers, or oligomers). Input the chromatographic data and detection data into calculation software to obtain parameters such as Charge and Equilibrium for the adsorption model.

[0105] c) Use the above experiments to fit the model parameters and obtain accurate parameters such as Charge, Equilibrium, Kinetic, and Shielding for each component.

[0106] 2. Based on the established chromatographic mechanism model, identify key influencing factors (i.e., the target influencing factors) and conduct wet experiments to verify them.

[0107] 3. Using the key influencing factors identified through model screening, and combining them with the key influencing factors and separation targets identified in step 2, set the ranges for each parameter and design a DoE experiment. For the experimental conditions designed in the DoE (i.e., the second chromatography conditions), replace the wet experiment with the chromatography mechanism model to predict the prediction results for the second chromatography conditions. Construct a statistical model based on the prediction results of the chromatography mechanism model.

[0108] 4. From the second chromatography conditions, select 3-5 experimental conditions (i.e., the third chromatography conditions) for wet experimental verification. The specific selection of experimental conditions can be determined based on the prediction results of the chromatography mechanism model in step 3, so that the target value (such as the polymer level) corresponding to the third chromatography condition is distributed at high, medium and low levels.

[0109] 5. The statistical model based on the mechanistic model is corrected using the wet experimental results; this is called combined modeling, resulting in the final tomographic prediction model. Specifically,

[0110] a) Extract the significant factor terms from the statistical model output by the chromatographic mechanism model (step 3).

[0111] b) Align the wet experimental results corresponding to the third chromatography condition with the prediction results of the chromatography mechanism model to obtain aligned data values. Specifically, linear regression or maximum-minimum normalization can be used for data alignment.

[0112] c) The prediction results of the tomographic mechanism model and the data-aligned wet experimental results are combined to construct a tomographic prediction model. Specifically, the significant factor term extracted in step 5a) and the data source "tomographic mechanism model" or "wet experiment" are added as a new influencing factor to construct the tomographic mechanism model.

[0113] Figure 4 This is a prediction effect diagram of a preliminary prediction model provided according to an embodiment of the present invention. Figure 5 This is a prediction effect diagram of a tomographic prediction model provided by an embodiment of the present invention. Based on the above embodiment, the following is combined with... Figure 4 and Figure 5 An optional embodiment of the method for determining the tomographic prediction model is described in detail.

[0114] 1. To construct a chromatographic mechanism model for monoclonal antibodies, with the goal of separating high-polymers.

[0115] 2. Based on the chromatography mechanism model, three key influencing factors (i.e., the target influencing factors) were identified, and their target values ​​and parameter ranges were set. For example, the sample loading capacity was: target value ± 23 g / L packing material, the elution pH was: target value ± 0.1, and the elution salt concentration was: target value ± 3 mM sodium chloride.

[0116] 3. Based on the target influencing factors and parameter ranges in step 2, a DoE experimental design was conducted. For the experimental conditions designed for DoE, a chromatography mechanism model was used to replace the wet experiment to obtain predicted results. A statistical model was established based on the predicted results of the chromatography mechanism model (specifically, the input factors are three experimental parameters; the response is the polymer result predicted by the chromatography mechanism model). The prediction results are shown in Table 1, with R² = 0.92. Where Intercept represents the intercept, Ep represents the elution pH, Ld represents the loading amount, and Es represents the elution salt concentration. The prediction effect of the preliminary prediction model constructed based on this is as follows: Figure 4As shown, the horizontal axis X represents the predicted value of the tomography mechanism model, and the vertical axis Y represents the predicted value of the preliminary prediction model. The red dashed line is the plot of Y=X. The blue dots in the figure represent the plot of the tomography mechanism model's prediction results for 18 experimental conditions as X and the preliminary prediction model's prediction results as Y. The data points are evenly distributed on both sides of the red dashed line and close to it, indicating that the preliminary prediction model can accurately predict the results of the tomography mechanism model. The prediction effect of the preliminary prediction model is shown in [the figure]. Figure 4 The p_value in Table 1 represents the degree of contribution of the target influencing factors to the chromatographic target item. When the p_value is less than or equal to 0.05, it indicates that the target influencing factors have a significant contribution to the chromatographic target item.

[0117] Table 1

[0118] term coefficient p_value Intercept -9.37972 3.52E-08 Ep 1.726981 2.18E-08 Ld×Es 0.000131 0.004334 Ld^2 -0.0001 0.020851

[0119] 4. Select five experimental conditions corresponding to high, medium and low polymer levels (i.e., the third chromatography conditions) and conduct offline wet experiments.

[0120] 5. Align the offline wet experiment results with the prediction results of the corresponding tomographic mechanism model, and then construct a tomographic prediction model based on the data alignment results and the preliminary prediction model. Specifically, using the significant factor extracted in step 3, and adding "tomographic mechanism model" or "wet experiment" as a new influencing factor, a tomographic mechanism model is constructed. The final tomographic prediction model is shown in Table 2, with R² = 0.97. The prediction effect of the tomographic prediction model is shown in [Table 2]. Figure 5 .

[0121] Table 2

[0122]

[0123] against Figure 5 The data within the green shaded area above is represented by the x-axis (X) of the wet experiment and the y-axis (Y) of the tomographic prediction model's prediction. Five green points correspond to the results of the five experimental conditions selected in step 4, while the other red points represent 13 additional wet experiment results (corresponding to the DoE experimental design conditions), used to assess the prediction performance of the final tomographic prediction model. All data points (green and red) are distributed on both sides of the red dashed line (X=Y) and are close to it, indicating that the final tomographic prediction model's prediction performance is accurate. Figure 5The data within the blue shaded area below represents the actual results of the wet experiment (x-axis) and the predicted values ​​of the initial prediction model directly constructed using the tomographic mechanism model (y-axis). The data points (gray) deviate significantly from the red dashed line (X=Y), indicating that the preliminary prediction model had poor predictive ability. The comparison of these results shows that the prediction ability of the final tomographic prediction model is significantly improved.

[0124] Based on the embodiments of this invention, a preliminary prediction model can be established using the prediction results of the tomographic mechanism model, and then corrected using offline wet experimental results. This invention fully utilizes the trend information obtained from the tomographic mechanism model and the absolute results of offline wet experiments to obtain a tomographic prediction model with higher prediction accuracy. When correcting the model using wet experimental results, aligning the prediction results of the model corresponding to the third tomographic condition with the actual results of the wet experiment before model correction can make the model prediction more accurate. Compared to prediction models built solely based on the tomographic mechanism model, the final tomographic prediction model predicts the absolute values ​​of the results more accurately. Compared to traditional prediction models built based on wet experiments, fewer offline wet experiments can be conducted, reducing the waste of experimental resources.

[0125] Example 3

[0126] Figure 6 This is a schematic diagram of a device for determining a tomographic prediction model provided in Embodiment 3 of the present invention. Figure 6 As shown, the device includes: a mechanism model determination module 310, an influencing factor screening module 320, a sample data determination module 330, and a prediction model construction module 340.

[0127] The system includes a mechanism model determination module 310 for acquiring a tomographic mechanism model under a given tomographic scenario; an influencing factor screening module 320 for determining a target influencing factor from multiple undetermined influencing factors under a given tomographic scenario based on the tomographic mechanism model; a sample data determination module 330 for acquiring first sample data obtained through the tomographic mechanism model for the target influencing factor; and acquiring second sample data obtained through wet experiments for the target influencing factor; the first sample data and the second sample data respectively include the data value of the target influencing factor and the data value of the tomographic target item; and a prediction model construction module 340 for constructing a tomographic prediction model based on the first sample data and the second sample data, wherein the tomographic prediction model characterizes the mapping relationship between the tomographic target item and the target influencing factor.

[0128] The technical solution of this invention involves: acquiring a tomographic mechanism model under a given tomographic scenario; determining a target influencing factor from multiple undetermined influencing factors under the given tomographic scenario based on the tomographic mechanism model; acquiring first sample data obtained through the tomographic mechanism model for the target influencing factor; and acquiring second sample data obtained through wet experiments for the target influencing factor. The first and second sample data respectively include data values ​​of the target influencing factor and data values ​​of the tomographic target item. A tomographic prediction model is constructed based on the first and second sample data, and the tomographic prediction model characterizes the mapping relationship between the tomographic target item and the target influencing factor. This solves the technical problem that the tomographic mechanism model is limited by computational solving capabilities, resulting in poor tomographic prediction accuracy. This invention achieves the effect of constructing a tomographic prediction model with higher accuracy by combining the tomographic prediction data of the tomographic mechanism model and the real experimental data of wet experiments.

[0129] Optionally, the influencing factor screening module 320 includes: an influence degree determination unit and an influencing factor screening unit;

[0130] The influence degree determination unit is used to determine the influence degree of each of the undetermined influencing factors on the tomographic target item under the set tomographic scenario based on the tomographic mechanism model.

[0131] The influencing factor screening unit is used to determine the target influencing factor among a plurality of undetermined influencing factors based on the degree of influence.

[0132] Optionally, the influence degree determination unit is configured to: determine multiple first chromatographic conditions corresponding to multiple undetermined influencing factors using statistical experimental design methods, each first chromatographic condition including data values ​​corresponding to multiple undetermined influencing factors; determine a first target value of the chromatographic target item under each first chromatographic condition based on the chromatographic mechanism model, and determine the influence degree of each undetermined influencing factor on the chromatographic target item according to the first target value of the chromatographic target item corresponding to different chromatographic conditions.

[0133] Optionally, the sample data determination module includes: a second condition determination unit, a second target value determination unit, and a first sample determination unit;

[0134] The second condition determination unit is used to determine multiple second stratification conditions corresponding to the target influencing factor, and each second stratification condition includes the data value of the target influencing factor.

[0135] The second target value determination unit is used to obtain the second target value of the tomographic target item obtained by the tomographic mechanism model for each of the second tomographic conditions;

[0136] The first sample determination unit is used to form the first sample data based on each of the second chromatography conditions and the second target value of the chromatography target item.

[0137] Optionally, the sample data determination module includes: a third condition determination unit, a third target value determination unit, and a second sample determination unit;

[0138] The third condition determination unit is used to classify the second stratification condition based on the second target value to obtain multiple data classes, and to determine the third stratification condition in the second stratification condition in each data class.

[0139] The third target value determination unit is used to obtain the third target value of the chromatography target item obtained by the wet test for each of the third chromatography conditions;

[0140] The second sample determination unit is used to form the second sample data based on each of the third chromatography conditions and the third target value of the chromatography target item.

[0141] Optionally, the prediction model building module includes: a preliminary model building unit, a source label setting unit, and a prediction model determination unit;

[0142] The preliminary model building unit is used to build a preliminary prediction model based on the first sample data. The preliminary prediction model represents the preliminary mapping relationship between the stratified target item and the target influencing factor.

[0143] The source label setting unit is used to set data source labels for the first sample data and the second sample data respectively;

[0144] The prediction model determination unit is used to adjust the preliminary prediction model based on the first sample data and the second sample data carrying the data source label to obtain a stratified prediction model. The stratified prediction model characterizes the mapping relationship between the stratified target item and the data source and the target influencing factors.

[0145] Optionally, the prediction model determination unit includes: a data alignment subunit and a preliminary model adjustment subunit;

[0146] The data alignment subunit is used to perform data alignment on the first sample data and the second sample data carrying the data source label to obtain aligned sample data. The aligned sample data includes the data source label, the data value of the target influencing factor, and the aligned data value of the chromatographic target item.

[0147] The preliminary model adjustment subunit is used to adjust the preliminary prediction model based on the aligned sample data, taking the data source as an additional influencing factor, to obtain the chromatographic prediction model.

[0148] The apparatus for determining the tomographic prediction model provided in the embodiments of the present invention can execute the method for determining the tomographic prediction model provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method.

[0149] Example 4

[0150] Figure 7 This is a flowchart of a tomographic data prediction method provided in Embodiment 4 of the present invention. This embodiment is applicable to situations where tomographic data prediction is performed using a tomographic prediction model. The method can be executed by a tomographic data prediction device, which can be implemented in hardware and / or software and can be configured in a computer. Figure 7 As shown, the method includes:

[0151] S410. Determine the target input values ​​for the target influencing factors in the tomography scenario.

[0152] The target input value can be understood as the input value of the chromatographic prediction model, or as the input value for each target influencing factor. For example, when the target influencing factors are sample loading, elution pH, and elution salt concentration, the target input value may include a first input value corresponding to the sample loading, a second input value corresponding to the elution pH, and a third input value corresponding to the elution salt concentration. In this embodiment of the invention, the input values ​​can be preset according to scenario requirements and are not specifically limited here. The first input value can be 20, 30, or 40 g / L. The second input value can be 3.2, 3.5, or 3.8, etc. The third input value can be 100, 150, or 200 mM sodium chloride, etc.

[0153] S420. Determine the target predicted value of the chromatographic target item determined by the chromatographic prediction model for the target influencing factors based on the target input value.

[0154] The target predicted value can be understood as the prediction result of the tomographic prediction model.

[0155] The technical solution of this invention involves determining the target input value of a target influencing factor under a defined tomographic scenario; and determining the target predicted value of the tomographic target item based on the target input value of the target influencing factor using a tomographic prediction model. This invention improves the accuracy of tomographic effect prediction under tomographic separation conditions using the constructed tomographic prediction model.

[0156] Example 5

[0157] Figure 8 This is a schematic diagram of a tomographic data prediction device provided in Embodiment 5 of the present invention. Figure 8 As shown, the device includes an input value determination module 510 and a tomographic prediction module 520.

[0158] The input value determination module 510 is used to determine the target input value of the target influencing factor under the set stratification scenario; the stratification prediction module 520 is used to determine the target prediction value of the stratification target item determined by the stratification prediction model for the target input value of the target influencing factor.

[0159] The technical solution of this invention involves determining the target input value of a target influencing factor under a defined tomographic scenario; and determining the target predicted value of the tomographic target item based on the target input value of the target influencing factor using a tomographic prediction model. This invention improves the accuracy of tomographic effect prediction under tomographic separation conditions using the constructed tomographic prediction model.

[0160] The chromatographic data prediction device provided in the embodiments of the present invention can execute the chromatographic data prediction method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0161] Example 6

[0162] Figure 9 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0163] like Figure 9As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0164] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0165] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for determining tomographic prediction models or tomographic data prediction methods.

[0166] In some embodiments, the method for determining a tomographic prediction model or the tomographic data prediction method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for determining a tomographic prediction model or the tomographic data prediction method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for determining a tomographic prediction model or the tomographic data prediction method by any other suitable means (e.g., by means of firmware).

[0167] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0168] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0169] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0170] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0171] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0172] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0173] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0174] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for determining a tomographic prediction model, characterized in that, include: Obtain the tomography mechanism model under the specified tomography scenario; Based on the tomography mechanism model, the target influencing factor is determined from multiple undetermined influencing factors in a given tomography scenario; The process involves obtaining first sample data for the target influencing factor through the chromatographic mechanism model; and obtaining second sample data for the target influencing factor through a wet experiment; the first sample data and the second sample data respectively include data values ​​of the target influencing factor and data values ​​of the chromatographic target item. A chromatographic prediction model is constructed based on the first sample data and the second sample data. The chromatographic prediction model represents the mapping relationship between the chromatographic target item and the target influencing factors.

2. The method according to claim 1, characterized in that, The process of determining the target influencing factor from multiple undetermined influencing factors in a given tomographic scenario based on the tomographic mechanism model includes: Based on the tomography mechanism model, determine the degree of influence of each of the undetermined influencing factors on the tomography target item under the given tomography scenario; The target influencing factor is determined from among the multiple undetermined influencing factors based on the degree of influence.

3. The method according to claim 2, characterized in that, The determination of the degree of influence of each of the undetermined influencing factors on the tomographic target item under the given tomographic scenario based on the tomographic mechanism model includes: Multiple first stratification conditions corresponding to multiple undetermined influencing factors are determined using statistical experimental design methods, and each first stratification condition includes data values ​​corresponding to multiple undetermined influencing factors. Based on the tomography mechanism model, the first target value of the tomography target item is determined under each first tomography condition, and the degree of influence of each undetermined influencing factor on the tomography target item is determined according to the first target value of the tomography target item corresponding to different tomography conditions.

4. The method according to claim 1, characterized in that, The acquisition of the first sample data obtained through the tomographic mechanism model for the target influencing factor includes: Multiple second-level conditions corresponding to the target influencing factor are determined, and each second-level condition includes the data value of the target influencing factor; Obtain the second target value of the tomographic target item obtained through the tomographic mechanism model for each of the second tomographic conditions; The first sample data is formed based on each of the second tomographic conditions and the second target value of the tomographic target item.

5. The method according to claims 3 and 4, characterized in that, The acquisition of the second sample data obtained through wet experiments targeting the influencing factors includes: Based on the second target value, the second stratification condition is classified to obtain multiple data classes, and a third stratification condition is determined in the second stratification condition of each data class. Obtain the third target value of the chromatography target item obtained through the wet experiment for each of the third chromatography conditions; The second sample data is formed based on each of the third chromatography conditions and the third target value of the chromatography target item.

6. The method according to claim 1, characterized in that, The construction of the tomographic prediction model based on the first sample data and the second sample data includes: A preliminary prediction model is constructed based on the first sample data. The preliminary prediction model represents the preliminary mapping relationship between the chromatographic target item and the target influencing factors. Set data source labels for the first sample data and the second sample data respectively; The preliminary prediction model is adjusted based on the first sample data and the second sample data carrying the data source label to obtain a stratified prediction model. The stratified prediction model represents the mapping relationship between the stratified target item and the data source and the target influencing factors.

7. The method according to claim 6, characterized in that, The step of adjusting the preliminary prediction model based on the first sample data and the second sample data carrying the data source label to obtain the tomographic prediction model includes: Data alignment is performed on the first sample data and the second sample data carrying the data source label to obtain aligned sample data, which includes the data source label, the data value of the target influencing factor, and the aligned data value of the chromatographic target item; Based on the preliminary prediction model, the data source is added as a new influencing factor, and the preliminary prediction model is adjusted based on the aligned sample data to obtain the chromatographic prediction model.

8. A method for predicting tomographic data, characterized in that, include: Determine the target input values ​​for the target influencing factors in the tomographic scenario; Determine the target predicted value of the chromatographic target item based on the target input value of the target influencing factor determined by the chromatographic prediction model.

9. A device for determining a tomographic prediction model, characterized in that, include: The mechanism model determination module is used to obtain the tomography mechanism model under the given tomography scenario; The influencing factor screening module is used to determine the target influencing factor from multiple undetermined influencing factors under a set chromatographic scenario based on the chromatographic mechanism model. The sample data determination module is used to acquire first sample data obtained through the chromatographic mechanism model for the target influencing factor; and to acquire second sample data obtained through wet experiment for the target influencing factor; the first sample data and the second sample data respectively include the data value of the target influencing factor and the data value of the chromatographic target item; The prediction model building module is used to build a stratified prediction model based on the first sample data and the second sample data. The stratified prediction model represents the mapping relationship between the stratified target item and the target influencing factors.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method for determining the tomographic prediction model according to any one of claims 1-8 and the tomographic data prediction method according to any one of claims 9.

Images