Quality control method and platform for NK cell preparation

Abstract

The invention discloses a quality control method and platform for NK cell preparation, and relates to the technical field of cell preparation.The method comprises the steps that first preset culture environment information, EZH2 inhibitor weight parts and IL-15 fusion protein weight parts are processed through an NKG2D expression quantity prediction model, NKG2D predicted expression quantity is generated, when the NKG2D predicted expression quantity is inconsistent with target NKG2D expression quantity, directional optimization is conducted, and the NKG2D predicted expression quantity is obtained; obtaining culture environment baseline information meeting the target NKG2D expression quantity; when the culture environment monitoring time sequence information is inconsistent with the culture environment baseline information, an NK cell preparation quality control early warning signal is generated. The technical problem that in the large-scale preparation process, due to the uncertainty of the culture environment, the immunocompetence of the NK cells possibly has large difference is solved, and the purposes of ensuring the stability and consistency of the environment in the preparation process through a real-time monitoring and automatic early warning mechanism and improving the immunocompetence of the NK cells are achieved. Therefore, the stability and the consistency of the immunocompetence of the NK cells are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of cell preparation, and in particular to a quality control method and platform for NK cell preparation. Background Art

[0002] With the widespread application of immunotherapy in the treatment of diseases such as cancer and viral infections, NK cells have gradually become a hot topic in cell therapy research due to their natural antiviral and anti-tumor abilities. NK cells can play a key role in immune defense by identifying and killing abnormal cells, such as virus-infected cells or tumor cells. However, existing NK cell culture technology faces many challenges, especially in the large-scale preparation process. How to ensure the activity and consistency of NK cells remains an urgent problem to be solved. Summary of the invention

[0003] The present application provides a quality control method and platform for NK cell preparation, which is used to solve the technical problem that the immune activity of NK cells may vary greatly due to the uncertainty of the culture environment during large-scale preparation.

[0004] In a first aspect of the present application, a quality control method for NK cell preparation is provided, the method comprising: receiving first preset culture environment information, EZH2 inhibitor weight parts and IL-15 fusion protein weight parts; processing the first preset culture environment information, the EZH2 inhibitor weight parts and the IL-15 fusion protein weight parts through an NKG2D expression prediction model to generate a predicted NKG2D expression amount; when the predicted NKG2D expression amount is inconsistent with a target NKG2D expression amount, performing a targeted optimization on the first preset culture environment information, the EZ H2 inhibitor weight parts and the IL-15 fusion protein weight parts to obtain culture environment baseline information that meets the target NKG2D expression amount; receiving culture environment monitoring timing information; when the culture environment monitoring timing information is inconsistent with the culture environment baseline information, generating a quality control warning signal for NK cell preparation.

[0005] In a second aspect of the present application, a quality control platform for NK cell preparation is provided, the platform comprising: a basic information receiving module, the basic information receiving module is used to receive first preset culture environment information, EZH2 inhibitor weight parts and IL-15 fusion protein weight parts; an NKG2D expression amount prediction module, the NKG2D expression amount prediction module is used to process the first preset culture environment information, the EZH2 inhibitor weight parts and the IL-15 fusion protein weight parts through an NKG2D expression amount prediction model to generate a predicted NKG2D expression amount; a directional optimization module, the directional optimization module is used to When the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein are directed to optimize to obtain the culture environment baseline information that meets the target NKG2D expression level; a culture environment monitoring information receiving module, the culture environment monitoring information receiving module is used to receive the culture environment monitoring timing information; a quality control warning signal generating module, the quality control warning signal generating module is used to generate a NK cell preparation quality control warning signal when the culture environment monitoring timing information is inconsistent with the culture environment baseline information.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The quality control method and platform for NK cell preparation provided in this application relate to the field of cell preparation technology. By receiving preset culture environment information, EZH2 inhibitor and IL-15 fusion protein weight portions, and combining the NKG2D expression prediction model, the NKG2D predicted expression is generated. When the predicted value is inconsistent with the target value, the culture conditions are adjusted by directional optimization to obtain the baseline information that meets the target NKG2D expression, and the culture environment monitoring time series information is received in real time. If it is inconsistent with the baseline information, a quality control warning signal is generated. It solves the technical problem that there may be large differences in NK cell immune activity due to the uncertainty of the culture environment during large-scale preparation, and realizes the technical effect of ensuring the stability and consistency of the environment during the preparation process through real-time monitoring and automated early warning mechanisms, thereby improving the stability and consistency of NK cell immune activity. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0009] Figure 1A schematic diagram of a quality control method for NK cell preparation provided in an embodiment of the present application;

[0010] Figure 2 A schematic diagram of a process for obtaining the culture environment baseline information that meets the target NKG2D expression amount through directional optimization in the quality control method for NK cell preparation provided in an embodiment of the present application;

[0011] Figure 3 Schematic diagram of the quality control platform structure for NK cell preparation provided in an embodiment of the present application.

[0012] Explanation of the accompanying drawings: basic information receiving module 11, NKG2D expression level prediction module 12, directional optimization module 13, culture environment monitoring information receiving module 14, quality control warning signal generation module 15. DETAILED DESCRIPTION

[0013] The present application provides a quality control method and platform for NK cell preparation, which is used to solve the technical problem that the immune activity of NK cells may vary greatly due to the uncertainty of the culture environment during large-scale preparation.

[0014] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0015] It should be noted that the terms "first", "second", etc. in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices.

[0016] Embodiment 1, as Figure 1 As shown, the present application provides a quality control method for NK cell preparation, the method comprising:

[0017] P10: receiving the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein.

[0018] Specifically, this step focuses on receiving and integrating a series of key parameters to ensure that NK cells can obtain the most optimized growth conditions during the culture process. First, the first preset culture environment information is received, which usually includes the setting of key environmental parameters during the culture process, such as temperature, humidity, pH value, oxygen concentration, etc. These parameters determine the growth environment of NK cells and ensure the optimal state of their biological activity and metabolic function. Culture environment information is a core parameter of cell culture, which directly affects the proliferation and activity of NK cells, especially the expression of their immune function.

[0019] Next, receive the weight portion of EZH2 inhibitor and the weight portion of IL-15 fusion protein. EZH2 inhibitor is an important molecular regulatory factor. By inhibiting the activity of EZH2 (histone methyltransferase), it can prevent the activity of NK cells from being inhibited by the immune microenvironment, thereby enhancing its ability to kill tumors and virus-infected cells. The weight portion of EZH2 inhibitor refers to the specific amount of the inhibitor added, which must be within a reasonable range to achieve the best biological effect.

[0020] Similarly, IL-15 fusion protein is an important immunomodulator that promotes NK cell activity and can enhance the proliferation and activation of NK cells. IL-15 itself is a cytokine, and the fusion protein form increases its stability and bioavailability, allowing NK cells to more effectively identify and eliminate diseased cells. The weight of IL-15 fusion protein here also refers to its usage. Too little may not activate sufficient immune response, while too much may lead to unnecessary side effects or waste of resources.

[0021] The information received by the system will provide basic data for subsequent NK cell expansion and functional enhancement, ensuring that in the subsequent culture process, the use of environmental parameters and chemical factors can be precisely controlled to promote the effective expression of key immune receptors such as NKG2D, and ultimately improve the therapeutic effect of NK cells.

[0022] P20: The first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein are processed by the NKG2D expression prediction model to generate a predicted NKG2D expression level.

[0023] Specifically, the NKG2D expression prediction model is used to process the culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein received from the above steps to generate the predicted NKG2D expression of NK cells. This step is a key technical link in the NK cell preparation process and directly determines the immune activity and quality of the cells.

[0024] First of all, NKG2D (natural killer cell activation receptor) is an important immune receptor that is widely present on the surface of NK cells and is responsible for identifying and clearing virus-infected cells and tumor cells. The higher the expression level of NKG2D, the stronger the killing activity of NK cells. Therefore, controlling and predicting the expression of NKG2D is a core step to ensure that NK cells have a strong immune effect.

[0025] In this step, the system will generate the predicted NKG2D expression of NK cells under the environment and conditions based on the input first preset culture environment information, i.e., the culture temperature, oxygen concentration, humidity and other environmental parameters, combined with the specific usage of the two key chemical regulatory factors, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein, through the pre-trained NKG2D expression prediction model, to provide an accurate reference for subsequent culture and adjustment. Among them, the NKG2D expression prediction model can be trained based on a large amount of experimental data through technical means such as machine learning or biosimulation, and can accurately predict the expression of NKG2D under different culture conditions.

[0026] Furthermore, in order to construct the NKG2D expression prediction model, step P20 of the embodiment of the present application further includes:

[0027] P21: Collecting NK cell culture log data, wherein the NK cell culture log data includes culture environment record timing information, EZH2 inhibitor weight record information, IL-15 fusion protein weight record information and NKG2D expression monitoring value, wherein the culture environment record timing information, the EZH2 inhibitor weight record information and the IL-15 fusion protein weight record information are control variables of the NK cell culture process; P22: Aggregating the NK cell culture log data through the culture environment record timing information, the EZH2 inhibitor weight record information and the IL-15 fusion protein weight record information to obtain several groups of NK cell culture log data, wherein In the embodiment, the several groups of NK cell culture log data have several culture environment fusion time series data, several EZH2 inhibitor weight fusion values, several IL-15 fusion protein weight fusion values, and several groups of NKG2D expression monitoring values; P23: traverse the several groups of NKG2D expression monitoring values ​​to perform central trend evaluation, and obtain several NKG2D expression monitoring true values; P24: use the several NKG2D expression monitoring true values ​​as supervision data, perform fitting training on the several culture environment fusion time series data, the several EZH2 inhibitor weight fusion values, and the several IL-15 fusion protein weight fusion values, and obtain the NKG2D expression prediction model.

[0028] It should be understood that the specific process of constructing the NKG2D expression prediction model can be, first, widely collect NK cell culture log data, which is the basis for building the model, not only including the culture environment record time series information (such as temperature, humidity, gas concentration and other parameters that change over time), but also recording the weight information of EZH2 inhibitor and IL-15 fusion protein, and the corresponding NKG2D expression monitoring value. These control variables and monitoring values ​​together constitute a comprehensive record of the NK cell culture process, providing rich materials for subsequent data analysis and model training.

[0029] Next, these log data are processed using data aggregation technology. By integrating data points under the same or similar culture conditions, several groups of NK cell culture log data are obtained. These data groups not only contain the fused culture environment time series data (reflecting the overall trend of the culture conditions), but also contain the weight fusion values ​​of EZH2 inhibitors and IL-15 fusion proteins (representing the average or typical dosage of these key factors during the culture process), as well as the corresponding NKG2D expression monitoring value group. This step effectively reduces data redundancy and improves the efficiency of subsequent analysis.

[0030] Subsequently, the plurality of groups of NKG2D expression monitoring values ​​are traversed, and the central tendency of these NKG2D expression monitoring values ​​is evaluated. By calculating statistics such as the mean, median or mode, a plurality of NKG2D expression monitoring true values ​​are obtained. These true values ​​represent the typical levels of NKG2D expression under different culture conditions, providing reliable supervision data for model training.

[0031] Finally, the several NKG2D expression monitoring true values ​​are used as supervision data, and the fusion values ​​of the culture environment fusion time series data, EZH2 inhibitor and IL-15 fusion protein obtained above are fitted and trained. Through supervised learning, the optimal mapping relationship between the input variables (i.e., culture environment and chemical regulatory factors) and the output results (NKG2D expression) is found, so as to construct a high-precision NKG2D expression prediction model. Fitting training usually uses machine learning algorithms such as linear regression, decision trees or neural networks to ensure that the model can adapt to different culture conditions and make accurate predictions. Finally, a NKG2D expression prediction model with high prediction accuracy is obtained. This model can accurately reflect the relationship between culture conditions and NKG2D expression, and provides strong support for quality control in the NK cell preparation process.

[0032] Furthermore, step P22 of the embodiment of the present application also includes:

[0033] P22-1: Through the user end, configure the environmental indicator deviation threshold, environmental indicator deviation variance threshold, EZH2 inhibitor weight deviation threshold, and IL-15 fusion protein weight deviation threshold; P22-2: When the environmental indicator deviation time series mean is less than or equal to the environmental indicator deviation threshold, and the environmental indicator deviation variance is less than or equal to the environmental indicator deviation variance threshold, it is deemed to meet the aggregation condition 1; P22-3: When the EZH2 inhibitor weight deviation is less than or equal to the EZH2 inhibitor weight deviation threshold, it is deemed to meet the aggregation condition 2; P22-4: When the IL-15 fusion protein weight deviation is less than or equal to the IL-15 fusion protein weight deviation threshold, it is deemed Aggregation condition 3 is satisfied; P22-5: the first NK cell culture log data and the second NK cell culture log data that simultaneously satisfy the aggregation condition 1, the aggregation condition 2 and the aggregation condition 3 are aggregated into a group, otherwise, aggregation is not performed; P22-6: when the aggregation analysis of all NK cell culture log data is completed, the several groups of NK cell culture log data are traversed, and the mean of the culture environment record time series information, the mean of the EZH2 inhibitor weight record information, and the mean of the IL-15 fusion protein weight record information are counted, and they are set as the several culture environment fusion time series data, the several EZH2 inhibitor weight fusion values, and the several IL-15 fusion protein weight fusion values.

[0034] Optionally, by defining multiple key aggregation conditions and combining thresholds configured by the user, the NK cell culture log data is effectively screened and aggregated to generate reliable culture environment information, fusion values ​​of EZH2 inhibitor weight parts and IL-15 fusion protein weight parts.

[0035] First, through the user end, professionals configure a series of deviation thresholds based on empirical data, including environmental indicator deviation threshold, environmental indicator deviation variance threshold, EZH2 inhibitor weight deviation threshold, and IL-15 fusion protein weight deviation threshold. These thresholds are the criteria for the system to make judgments in subsequent data aggregation. The deviation threshold and variance threshold are used to limit the fluctuation range of environmental conditions, while the weight deviation thresholds of EZH2 inhibitors and IL-15 fusion proteins are used to ensure that the amount of chemical substances used fluctuates within a reasonable range to maintain the controllability and consistency of the NK cell culture process.

[0036] Next, the NK cell culture log data are checked one by one according to the preset deviation threshold. First, the deviation time series mean and variance of environmental indicators (such as temperature, humidity, etc.) are calculated and compared with the environmental indicator deviation threshold and the environmental indicator deviation variance threshold. Aggregation condition 1 is considered to be met only when the deviation time series mean of the environmental indicator is less than or equal to the environmental indicator deviation threshold and the deviation variance is less than or equal to the environmental indicator deviation variance threshold. This means that during the entire culture process, the environmental fluctuations are within an acceptable range and there is no excessive deviation.

[0037] Next, the system checks whether the EZH2 inhibitor weight deviation is within the defined threshold range. If the deviation is less than or equal to the configured EZH2 inhibitor weight deviation threshold, aggregation condition 2 is met. This step ensures that the amount of EZH2 inhibitor used meets the requirements, thereby maintaining the immunomodulatory effect of NK cells.

[0038] Similarly, the system will evaluate the deviation of the weight of IL-15 fusion protein, and when its deviation is less than or equal to the deviation threshold of the weight of IL-15 fusion protein, it is considered to meet the aggregation condition 3. This ensures that the dosage of IL-15 fusion protein is accurately controlled to promote NK cell activation.

[0039] Furthermore, the above three aggregation conditions are comprehensively evaluated. Only when the three conditions are met at the same time, that is, when aggregation condition 1, aggregation condition 2 and aggregation condition 3 are met at the same time, the currently investigated NK cell culture log data (such as the first NK cell culture log data and the second NK cell culture log data) will be regarded as similar data and aggregated into a group. If any one of the conditions is not met, no aggregation operation is performed. Such an aggregation strategy ensures that the aggregated data group has a high degree of similarity and consistency in culture conditions.

[0040] Finally, after all NK cell culture log data have been aggregated and analyzed, the system will traverse the aggregated data groups and calculate the mean of the culture environment record time series information, the EZH2 inhibitor weight record information, and the IL-15 fusion protein weight record information for each group of data. These mean data will be used as feature variable inputs for subsequent model training, namely, several culture environment fusion time series data, several EZH2 inhibitor weight fusion values, and several IL-15 fusion protein weight fusion values. This step not only simplifies the data representation, but also improves the efficiency and accuracy of model training.

[0041] P30: When the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, a targeted optimization is performed on the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein to obtain the culture environment baseline information that meets the target NKG2D expression level.

[0042] Specifically, when the system detects that the predicted NKG2D expression fails to reach the set target NKG2D expression, it means that the current culture conditions cannot ensure the optimal immune activity of NK cells. At this time, in view of the inconsistency between the predicted NKG2D expression and the target NKG2D expression, a directional optimization operation is performed to adjust the culture conditions to obtain the culture environment baseline information that meets the target NKG2D expression. Among them, the target NKG2D expression is the expected expression of the NKG2D, which can ensure the optimal immune activity of NK cells. This step is the core link in optimizing the key parameters in the NK cell culture process, and the immune activity of NK cells is ensured by accurately adjusting the culture environment and regulatory factors.

[0043] Further, when the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, step P30 of the embodiment of the present application further includes:

[0044] P31: When the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is greater than the NKG2D expression amount deviation threshold, the predicted NKG2D expression amount is deemed to be inconsistent with the target NKG2D expression amount, and the NKG2D expression amount deviation threshold is a user-side pre-configured parameter; P32: When the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is less than or equal to the NKG2D expression amount deviation threshold, the predicted NKG2D expression amount is deemed to be consistent with the target NKG2D expression amount.

[0045] In a possible embodiment of the present application, in order to further ensure the consistency between the predicted NKG2D expression and the target NKG2D expression, an evaluation mechanism for the NKG2D expression deviation between the two is introduced. By setting the NKG2D expression deviation threshold, the system can more accurately determine whether the culture conditions need to be optimized and adjusted.

[0046] First, after the NKG2D predicted expression is calculated, it is compared with the set target NKG2D expression to calculate the NKG2D expression deviation between the two. If the deviation is greater than the NKG2D expression deviation threshold pre-configured by the user, it is considered that the NKG2D predicted expression is inconsistent with the target NKG2D expression. At this point, the system needs to start a directional optimization process to adjust the culture conditions to reduce the deviation and reach the target expression.

[0047] Among them, the NKG2D expression deviation threshold is a preconfigured parameter set by the user according to experimental or production requirements. This parameter is used to define the allowable NKG2D expression fluctuation range. Fluctuations within this range are considered acceptable and do not require further optimization. Once the deviation exceeds the threshold, the system will identify that optimization is required.

[0048] If the calculated NKG2D expression deviation is less than or equal to the preset NKG2D expression deviation threshold, it can be considered that the predicted NKG2D expression has reached the target NKG2D expression requirement. This means that the current culture conditions are ideal and no further adjustments are required. The system will continue to amplify cells according to the culture conditions to ensure that NK cells continue to grow under standard conditions. In this case, the current culture environment information, EZH2 inhibitor weight parts, and IL-15 fusion protein weight parts are saved as standard values ​​as a reference for subsequent batch cultures. This mechanism not only ensures the efficiency of the current culture process, but also provides reliable reference data for future production.

[0049] By setting the NKG2D expression deviation threshold, the system introduces a more refined control method to ensure that the monitoring and adjustment of NKG2D expression is within a reasonable range during each culture process. As an important indicator for judging the culture effect, the NKG2D expression deviation can help the system determine in real time whether the culture conditions need to be adjusted to maintain the immune activity and function of NK cells. This deviation control mechanism not only improves the intelligence level of the system, but also ensures the consistency and stability of the product during the production process.

[0050] In practical applications, users can configure the NKG2D expression deviation threshold according to specific experimental requirements or product specifications to adapt to different culture environments and cell types and achieve personalized quality control.

[0051] Further, such as Figure 2 As shown, the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein are subjected to directional optimization, and step P30 of the embodiment of the present application further includes:

[0052] P33: Through the user end, the first preset culture environment information, the EZH2 inhibitor weight and the IL-15 fusion protein weight are adjusted to construct a plurality of culture control particles, wherein any one of the culture control particles includes the second preset culture environment information, the EZH2 inhibitor weight and the IL-15 fusion protein weight; P34: Through the NKG2D expression prediction model, the second preset culture environment information, the EZH2 inhibitor weight and the IL-15 fusion protein weight of the plurality of culture control particles are traversed for processing to obtain a plurality of NKG2D predicted expression amounts; P35: According to the plurality of NKG2D predicted expression amounts, the plurality of culture control particles are directed to optimize to obtain recommended culture control particles that theoretically meet the target NKG2D expression amount; P36: The recommended culture control particles are sent to the user end to obtain user feedback information; P37: When the user feedback information is passed, the culture environment baseline information is configured according to the recommended culture control particles.

[0053] Optionally, the culture conditions are adjusted by targeted optimization, using a variety of possible combinations of culture conditions to find the best solution to ensure that the NKG2D expression of NK cells reaches the target level.

[0054] First, using the user-side platform as an interface, the user or experimenter is allowed to adjust the first preset culture environment information (including but not limited to temperature, humidity, pH value, gas concentration, etc.), the weight of the EZH2 inhibitor, and the weight of the IL-15 fusion protein according to preliminary experimental data and theoretical experience, and encapsulate the adjusted parameter combination into a number of culture control particles, each of which represents a specific culture condition combination, including the second preset culture environment information (i.e., the adjusted environmental conditions, such as temperature, oxygen concentration, etc.), the preset weight of the EZH2 inhibitor, and the preset weight of the IL-15 fusion protein. By constructing multiple different particles, the system can traverse different culture conditions to find the optimal solution.

[0055] Subsequently, a pre-trained NKG2D expression prediction model is introduced. This model is based on machine learning or deep learning algorithms, and can quickly and accurately predict the expression of NKG2D under different culture conditions. Through the constructed NKG2D expression prediction model, all constructed culture control particles are traversed. The second preset culture environment information, the preset weight of the EZH2 inhibitor, and the preset weight of the IL-15 fusion protein in each particle will be input into the prediction model, and the corresponding NKG2D predicted expression will be generated according to these input parameters. This traversal process allows the system to evaluate the changes in the expression of NKG2D under different combinations of conditions, thereby providing a basis for subsequent optimization.

[0056] Furthermore, all culture control particles are optimized according to the predicted NKG2D expression level corresponding to each particle. The purpose of the targeted optimization is to find the recommended culture control particle closest to the target NKG2D expression level by calculating the deviation between the predicted NKG2D expression level of each particle and the target expression level. This process is the core step for the system to optimize according to the performance of each particle, ensuring that the output culture conditions can maximize the immune effect of NK cells.

[0057] When the theoretically optimal recommended culture control particles are found, these particles and related information are sent to the user for confirmation. And based on the user's actual needs and conditions, wait for user feedback. Users can evaluate the recommended culture conditions based on experimental results or production requirements to ensure their practical feasibility. If the user feedback information is passed, it means that the plan is accepted and ready to be implemented.

[0058] Finally, when the user feedback information is passed, the baseline information of the NK cell culture environment is automatically configured according to the recommended culture control particles confirmed by the user. This process may involve adjusting the settings of the incubator, accurately weighing and adding EZH2 inhibitors and IL-15 fusion proteins, etc., to ensure that NK cells are cultured under optimal conditions, thereby maximizing the expression of NKG2D and enhancing the immune efficacy of NK cells.

[0059] Furthermore, step P35 of the embodiment of the present application also includes:

[0060] P35-1: According to the predicted expression levels of the NKG2Ds, the better particles are sorted for the culture control particles to obtain k directional control particles, where k is an integer and k≥5; P35-2: According to the predicted expression levels of the NKG2Ds, the worse particles are sorted for the culture control particles to obtain j directional control particles, where j is an integer and j≥15 and j≥3*k; P35-3: Taking the k directional control particles as adjustment targets, the j directional control particles are crossovered and mutated to obtain updated culture control particles; P35-4: According to the NKG2D expression level prediction model, the recommended culture control particles that meet the target NKG2D expression level are sorted from the updated culture control particles.

[0061] It should be understood that the culture control particles are subjected to crossover, mutation and other operations through sorting and evolutionary algorithms to find the optimal combination of culture conditions to ensure that the NKG2D expression of NK cells reaches the target level.

[0062] First, all the culture control particles are preliminarily screened according to the predicted expression levels of the several NKG2Ds, and the first k particles with higher predicted expression levels of NKG2D are selected as the better particles. The predicted expression levels of NKG2D of these better particles are closest to the target NKG2D expression level, so they are regarded as better particles. k is an integer, and k≥5, ensuring that at least 5 best particles enter the next round of operation. This step is similar to the elite selection in the genetic algorithm, that is, retaining the best performing individuals as the basis for subsequent optimization.

[0063] At the same time, the inferior particles are sorted according to the NKG2D predicted expression, and j directional control particles with relatively poor performance are selected. The NKG2D predicted expression of these particles deviates far from the target value, so further optimization is required. j is an integer, and j≥15, and the condition of j≥3*k is satisfied, ensuring that the number of inferior particles is more than three times that of the superior particles. By selecting inferior particles, a large number of potential improvement objects are retained, increasing the optimization space.

[0064] Furthermore, taking k directional control particles as the adjustment targets, j directional control particles are updated through crossover and mutation operations. Crossover refers to combining some parameters of better particles with some parameters of worse particles to generate new particles, while mutation introduces small-scale random changes in certain parameters to increase diversity. This method is similar to the crossover and mutation operations in genetic algorithms, which aims to expand the search space and increase the possibility of finding the optimal solution. Through these two operations, a new batch of updated cultivation control particles are generated, which combine some advantages of better particles and introduce new possibilities through mutation.

[0065] After generating new updated culture control particles, the NKG2D expression prediction model is used again to evaluate these particles and calculate the predicted NKG2D expression of each particle. Based on these prediction results, the recommended culture control particles that meet the target NKG2D expression are selected. These particles have the culture conditions that are closest to or reach the target NKG2D expression, and become the optimal solution recommended to the user.

[0066] Furthermore, step P35 of the embodiment of the present application also includes:

[0067] If the recommended culture control particles that meet the target NKG2D expression level are not available, the culture control particles are iteratively updated.

[0068] Optionally, during the initial screening of recommended culture control particles, if all generated particles fail to meet the target NKG2D expression level, the iterative update process will automatically begin, which means that the existing culture condition combination is not sufficient to achieve the expected NK cell immune activity and the system needs to be further optimized.

[0069] In the iterative update, the system will retain the current optimal k directional control particles, and continue to perform crossover and mutation operations on these particles to generate new particles. Each iteration will try to introduce more variable changes or strengthen the existing optimal conditions based on the results of the previous round of optimization. And in each iteration, all generated updated culture control particles are re-evaluated, and the new predicted expression is calculated through the NKG2D expression prediction model. New solutions are generated through crossover and mutation operations, and gradually tend to better solutions until the culture conditions that can meet the target NKG2D expression are found.

[0070] The iterative update process is usually repeated many times until a particle combination that meets the requirements is found. Each iteration will further optimize based on the previous results, so this process requires strong automation support. Through the cooperation of the system's prediction model and optimization algorithm, the iteration can run automatically without human intervention until the optimal solution is found.

[0071] If the particle combination that meets the goal cannot be found after multiple rounds of iterations, the optimization efficiency can be further improved by adjusting certain basic parameters, such as expanding the search space or increasing the amplitude of random mutation.

[0072] P40: Receive the timing information of culture environment monitoring.

[0073] Optionally, during the NK cell preparation process, the system collects real-time culture environment monitoring time series information through sensors or monitoring equipment. This information includes key environmental parameters such as temperature, humidity, oxygen concentration, carbon dioxide concentration, pH value, etc., which directly affect the growth and activity of NK cells. Time series information refers to the dynamic changes of these parameters during the culture process, which can help the system continuously track changes in environmental conditions.

[0074] By receiving the culture environment monitoring time series information, the system can ensure real-time control of the culture environment and promptly discover any adverse factors that may affect the growth of NK cells. This step is the basis of the culture process, ensuring the real-time and accuracy of the data, and providing necessary support for subsequent comparison and early warning mechanisms.

[0075] P50: When the culture environment monitoring timing information is inconsistent with the culture environment baseline information, a NK cell preparation quality control warning signal is generated.

[0076] Specifically, after receiving the culture environment monitoring time series information, the system will compare it with the culture environment baseline information obtained by previous optimization. The culture environment baseline information is the optimal culture condition obtained through previous optimization and experiments, including the ideal range of all environmental parameters. If the current culture environment monitoring time series information is inconsistent with the culture environment baseline information, it means that the culture environment has deviated.

[0077] If the system finds inconsistencies between the monitoring timing information and the culture environment baseline information, that is, one or more key parameters deviate from the preset standard range, it will immediately trigger an early warning signal for NK cell preparation quality control. This early warning signal is the system's rapid response to potential problems, designed to remind operators to pay attention and take appropriate corrective measures in a timely manner. The generation of early warning signals is based on advanced algorithms and logical judgments, which can accurately identify abnormal conditions in the culture environment and evaluate their possible impact on the quality of NK cell preparation. This signal can prevent adverse culture environments from having a negative impact on NK cells in a timely manner, thereby ensuring the quality and stability of cell preparation.

[0078] In summary, the embodiments of the present application have at least the following technical effects:

[0079] The present application generates a predicted NKG2D expression by receiving preset culture environment information, EZH2 inhibitor and IL-15 fusion protein weight portions, and combining with the NKG2D expression prediction model. When the predicted value is inconsistent with the target value, the culture conditions are adjusted through targeted optimization to obtain baseline information that meets the target NKG2D expression, and the culture environment monitoring time series information is received in real time. If it is inconsistent with the baseline information, a quality control warning signal is generated.

[0080] The technical effect of ensuring the stability and consistency of the environment during the preparation process through real-time monitoring and automated early warning mechanisms has been achieved, thereby improving the stability and consistency of NK cell immune activity.

[0081] Example 2, based on the same inventive concept as the quality control method for NK cell preparation in the previous example, Figure 3 As shown, the present application provides a quality control platform for NK cell preparation, and the platform and method embodiments in the present application embodiments are based on the same inventive concept. Among them, the platform includes:

[0082] The basic information receiving module 11 is used to receive the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein.

[0083] The NKG2D expression prediction module 12 is used to process the first preset culture environment information, the EZH2 inhibitor weight and the IL-15 fusion protein weight through the NKG2D expression prediction model to generate the NKG2D predicted expression level.

[0084] The directional optimization module 13 is used to perform directional optimization on the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein when the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, so as to obtain the culture environment baseline information that meets the target NKG2D expression level.

[0085] The culture environment monitoring information receiving module 14 is used to receive culture environment monitoring time sequence information.

[0086] The quality control warning signal generating module 15 is used to generate a quality control warning signal for NK cell preparation when the culture environment monitoring timing information is inconsistent with the culture environment baseline information.

[0087] Furthermore, the NKG2D expression prediction module 12 is also used to perform the following steps:

[0088] Collect NK cell culture log data, wherein the NK cell culture log data includes culture environment record time sequence information, EZH2 inhibitor weight record information, IL-15 fusion protein weight record information and NKG2D expression monitoring value, the culture environment record time sequence information, the EZH2 inhibitor weight record information, and the IL-15 fusion protein weight record information are control variables of the NK cell culture process; perform aggregation on the NK cell culture log data through the culture environment record time sequence information, the EZH2 inhibitor weight record information, and the IL-15 fusion protein weight record information to obtain several groups of NK cell culture log data, wherein In the method, the several groups of NK cell culture log data have several culture environment fusion time series data, several EZH2 inhibitor weight fusion values, several IL-15 fusion protein weight fusion values, and several groups of NKG2D expression monitoring values; the several groups of NKG2D expression monitoring values ​​are traversed to perform central trend evaluation to obtain several NKG2D expression monitoring true values; using the several NKG2D expression monitoring true values ​​as supervision data, fitting training is performed on the several culture environment fusion time series data, the several EZH2 inhibitor weight fusion values, and the several IL-15 fusion protein weight fusion values ​​to obtain the NKG2D expression prediction model.

[0089] Furthermore, the NKG2D expression prediction module 12 is also used to perform the following steps:

[0090] Through the user end, configure the environmental indicator deviation threshold, environmental indicator deviation variance threshold, EZH2 inhibitor weight deviation threshold, and IL-15 fusion protein weight deviation threshold; when the environmental indicator deviation time series mean is less than or equal to the environmental indicator deviation threshold, and the environmental indicator deviation variance is less than or equal to the environmental indicator deviation variance threshold, it is considered that the aggregation condition 1 is met; when the EZH2 inhibitor weight deviation is less than or equal to the EZH2 inhibitor weight deviation threshold, it is considered that the aggregation condition 2 is met; when the IL-15 fusion protein weight deviation is less than or equal to the IL-15 fusion protein weight deviation threshold, it is considered that the aggregation condition 1 is met. 3; aggregate the first NK cell culture log data and the second NK cell culture log data that simultaneously meet the aggregation condition 1, the aggregation condition 2 and the aggregation condition 3 into a group, otherwise, do not perform aggregation; when the aggregation analysis of all NK cell culture log data is completed, traverse the several groups of NK cell culture log data, and count the mean of the culture environment record time series information, the mean of the EZH2 inhibitor weight record information, and the mean of the IL-15 fusion protein weight record information, and set them as the several culture environment fusion time series data, the several EZH2 inhibitor weight fusion values, and the several IL-15 fusion protein weight fusion values.

[0091] Furthermore, the directional optimization module 13 is also used to perform the following steps:

[0092] When the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is greater than the NKG2D expression amount deviation threshold, the predicted NKG2D expression amount is deemed to be inconsistent with the target NKG2D expression amount, and the NKG2D expression amount deviation threshold is a user-side pre-configured parameter; when the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is less than or equal to the NKG2D expression amount deviation threshold, the predicted NKG2D expression amount is deemed to be consistent with the target NKG2D expression amount.

[0093] Furthermore, the directional optimization module 13 is also used to perform the following steps:

[0094] Through the user end, the first preset culture environment information, the EZH2 inhibitor weight portion and the IL-15 fusion protein weight portion are adjusted to construct a plurality of culture control particles, wherein any one of the culture control particles includes the second preset culture environment information, the EZH 2 inhibitor weight portion and the IL-15 fusion protein weight portion; through the NKG2D expression amount prediction model, the second preset culture environment information, the EZH2 inhibitor weight portion and the IL-15 fusion protein weight portion of the plurality of culture control particles are traversed for processing to obtain a plurality of NKG2D predicted expression amounts; according to the plurality of NKG2D predicted expression amounts, the plurality of culture control particles are directed optimized to obtain recommended culture control particles that theoretically meet the target NKG2D expression amount; the recommended culture control particles are sent to the user end to obtain user feedback information; when the user feedback information is passed, the culture environment baseline information is configured according to the recommended culture control particles.

[0095] Furthermore, the directional optimization module 13 is also used to perform the following steps:

[0096] According to the predicted expression amounts of the NKG2Ds, the better particles are sorted for the culture control particles to obtain k directional control particles, where k is an integer and k≥5; according to the predicted expression amounts of the NKG2Ds, the worse particles are sorted for the culture control particles to obtain j directional control particles, where j is an integer and j≥15 and j≥3*k; with the k directional control particles as adjustment targets, the j directional control particles are crossovered and mutated to obtain updated culture control particles; according to the NKG2D expression amount prediction model, the recommended culture control particles that meet the target NKG2D expression amount are sorted from the updated culture control particles.

[0097] Furthermore, the directional optimization module 13 is also used to perform the following steps:

[0098] If the recommended culture control particles that meet the target NKG2D expression level are not available, the culture control particles are iteratively updated.

[0099] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0100] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0101] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A quality control method for NK cell preparation, characterized in that: include: Receiving first preset culture environment information, EZH2 inhibitor weight parts, and IL-15 fusion protein weight parts; The first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein are processed by a NKG2D expression prediction model to generate a predicted NKG2D expression level; When the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, performing targeted optimization on the first preset culture environment information, the weight of the EZH2 inhibitor, and the weight of the IL-15 fusion protein to obtain the culture environment baseline information that meets the target NKG2D expression level; Receive the time series information of culture environment monitoring; When the culture environment monitoring time series information is inconsistent with the culture environment baseline information, an NK cell preparation quality control early warning signal is generated.

2. The method according to claim 1, characterized in that The method of processing the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein by a NKG2D expression prediction model to generate a predicted NKG2D expression amount includes: Collecting NK cell culture log data, wherein the NK cell culture log data includes culture environment record time sequence information, EZH2 inhibitor weight record information, IL-15 fusion protein weight record information and NKG2D expression monitoring value, and the culture environment record time sequence information, the EZH2 inhibitor weight record information, and the IL-15 fusion protein weight record information are control variables of the NK cell culture process; Aggregating the NK cell culture log data by recording the culture environment timing information, the EZH2 inhibitor weight portion recording information, and the IL-15 fusion protein weight portion recording information to obtain several groups of NK cell culture log data, wherein the several groups of NK cell culture log data have several culture environment fusion timing data, several EZH2 inhibitor weight portion fusion values, several IL-15 fusion protein weight portion fusion values, and several groups of NKG2D expression monitoring values; Traversing the plurality of groups of NKG2D expression monitoring values ​​to perform central tendency evaluation, and obtaining a plurality of NKG2D expression monitoring true values; The plurality of NKG2D expression monitoring true values ​​are used as supervisory data, and fitting training is performed on the plurality of culture environment fusion time series data, the plurality of EZH2 inhibitor weight fusion values, and the plurality of IL-15 fusion protein weight fusion values ​​to obtain the NK G2D expression prediction model.

3. The method according to claim 2, characterized in that The NK cell culture log data is aggregated by recording the culture environment time sequence information, the EZH2 inhibitor weight portion recording information, and the IL-15 fusion protein weight portion recording information to obtain several groups of NK cell culture log data, including: Through the user end, configure the environmental indicator deviation threshold, environmental indicator deviation variance threshold, EZH2 inhibitor weight deviation threshold, and IL-15 fusion protein weight deviation threshold; When the time series mean of the environmental indicator deviation is less than or equal to the environmental indicator deviation threshold, and the environmental indicator deviation variance is less than or equal to the environmental indicator deviation variance threshold, it is deemed that aggregation condition 1 is met; When the weight deviation of the EZH2 inhibitor is less than or equal to the weight deviation threshold of the EZH2 inhibitor, it is considered that polymerization condition 2 is met; When the weight deviation of the IL-15 fusion protein is less than or equal to the weight deviation threshold of the IL-15 fusion protein, it is considered that the polymerization condition 3 is met; Aggregate the first NK cell culture log data and the second NK cell culture log data that simultaneously meet the aggregation condition 1, the aggregation condition 2, and the aggregation condition 3 into a group, otherwise, do not perform aggregation; After the aggregation analysis of all NK cell culture log data is completed, the several groups of NK cell culture log data are traversed, and the mean of the culture environment record time series information, the mean of the EZH2 inhibitor weight record information, and the mean of the IL-15 fusion protein weight record information are counted, and they are set as the several culture environment fusion time series data, the several EZH2 inhibitor weight fusion values, and the several IL-15 fusion protein weight fusion values.

4. The method according to claim 1, characterized in that When the predicted NKG2D expression level is inconsistent with the target NKG2D expression level, it includes: When the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is greater than the NKG2D expression amount deviation threshold, it is considered that the predicted NKG2D expression amount is inconsistent with the target NKG2D expression amount, and the NKG2D expression amount deviation threshold is a user-side preconfigured parameter; When the NKG2D expression amount deviation between the predicted NKG2D expression amount and the target NKG2D expression amount is less than or equal to the NKG2D expression amount deviation threshold, the predicted NKG2D expression amount is deemed to be consistent with the target NKG2D expression amount.

5. The method according to claim 1, characterized in that Directed optimization is performed on the first preset culture environment information, the weight of the EZH2 inhibitor, and the weight of the IL-15 fusion protein to obtain the culture environment baseline information that meets the target NKG2D expression amount, including: Through the user terminal, the first preset culture environment information, the EZH2 inhibitor weight portion and the IL-15 fusion protein weight portion are adjusted to construct a plurality of culture control particles, wherein any one of the culture control particles includes the second preset culture environment information, the EZH 2 inhibitor preset weight portion and the IL-15 fusion protein preset weight portion; By using the NKG2D expression amount prediction model, the second preset culture environment information of the plurality of culture control particles, the preset weight portions of the EZH2 inhibitor, and the preset weight portions of the IL-15 fusion protein are traversed for processing to obtain a plurality of NKG2D predicted expression amounts; According to the predicted expression amounts of the plurality of NKG2Ds, a directed optimization is performed on the plurality of culture control particles to obtain recommended culture control particles that theoretically meet the target NKG2D expression amount; Sending the recommended cultivation control particles to the user end to obtain user feedback information; When the user feedback information is passed, the cultivation environment baseline information is configured according to the recommended cultivation control particle.

6. The method according to claim 5, characterized in that According to the predicted expression amounts of the plurality of NKG2Ds, a directional optimization is performed on the plurality of culture control particles to obtain recommended culture control particles that theoretically meet the target NKG2D expression amount, including: According to the predicted expression amounts of the NKG2D, the plurality of cultured controlled particles are sorted to obtain k directional controlled particles, where k is an integer and k≥5; According to the predicted expression amounts of the NKG2D, the plurality of cultured controlled particles are sorted for inferior particles to obtain j directional controlled particles, where j is an integer, j≥15, j≥3*k; Taking the k directional control particles as adjustment targets, crossover and mutation are performed on the j directional control particles to obtain updated culture control particles; According to the NKG2D expression amount prediction model, the recommended culture control particles that meet the target NKG2D expression amount are sorted from the updated culture control particles.

7. The method according to claim 6, characterized in that include: If the recommended culture control particles that meet the target NKG2D expression level are not available, the culture control particles are iteratively updated.

8. A quality control system for NK cell preparation, characterized in that: The system comprises: A basic information receiving module, the basic information receiving module is used to receive the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein; An NKG2D expression amount prediction module, the NKG2D expression amount prediction module is used to process the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein through an NKG2D expression amount prediction model to generate a predicted NKG2D expression amount; A directional optimization module, wherein when the predicted NKG2D expression amount is inconsistent with the target NKG2D expression amount, the directional optimization module is used to perform directional optimization on the first preset culture environment information, the weight of the EZH2 inhibitor and the weight of the IL-15 fusion protein to obtain the culture environment baseline information that meets the target NKG2D expression amount; A culture environment monitoring information receiving module, wherein the culture environment monitoring information receiving module is used to receive culture environment monitoring time sequence information; A quality control warning signal generation module is used to generate a quality control warning signal for NK cell preparation when the culture environment monitoring timing information is inconsistent with the culture environment baseline information.

Images