Automatic material supplementing method based on real-time dynamic detection result

Abstract

The invention provides an automatic material supplementing method based on a real-time dynamic detection result, which comprises the following steps: (1) collecting the real-time dynamic detection result, and calculating the consumption rate and the generation rate of each component in unit time; selecting the component with the fastest consumption rate as a marker of the current time period; (2) collecting historical detection data, and calculating a cell growth rate in a logarithmic phase and a target product expression rate in a culture harvesting stage; taking the concentration level of the marker when the cell growth rate parameter in the logarithmic phase is maximum as a component concentration set value of the current time period of the logarithmic phase; taking the concentration level of the marker when the target product expression rate parameter of the culture harvesting stage is maximum as a component concentration set value of the current time period of the culture harvesting stage; and (3) calculating the additive amount of the marker in the current time period according to the consumption rate and the generation rate of the marker, and calculating the additive amount of the culture medium according to the additive amount of the marker to obtain a feeding strategy.

Description

Technical Field

[0001] The present invention belongs to the field of biotechnology, and particularly relates to an automated feeding method based on real-time dynamic detection results. Background Art

[0002] Cell culture has currently been widely used in the research and production of biopharmaceuticals. A crucial factor in ensuring the success of cell culture is the cell culture process. The quality of the cell culture process largely depends on the control of the culture conditions on the bioreactor. In addition to maintaining important physicochemical conditions such as pH, temperature, and dissolved oxygen (DO), meeting the nutritional requirements of cells plays an important role in establishing a suitable growth environment. The metabolism of mammalian cells involves complex metabolic pathways, and cells thus consume a large amount of nutrients, such as carbon sources, nitrogen sources, and trace elements. Currently, comprehensive feeding media are commonly used in the industry. This medium combines nutrients such as carbon sources, nitrogen sources, and trace elements according to the metabolic requirements of cells to form a comprehensive feeding medium.

[0003] At the present stage, during the cell culture process, in order to maintain the nutrients required for the cell growth environment, it is necessary to perform a feeding operation on the cell culture medium. However, the determination of the addition amount of the feeding medium and the feeding time point is often based on the experience of the existing process, which may lead to phenomena such as insufficient feeding amount added per unit time, excessive feeding amount added, and chaotic feeding time points, thus affecting process development and ultimately resulting in changes in batch-to-batch production.

[0004] The cell culture medium is a crucial influencing factor for the success of cell culture, and it has an important impact on how to regulate cell culture process parameters. The research and development of biopharmaceuticals require a large number of cell culture experiments. The most reliable cell culture process flow is screened through the experimental results to ensure a long-term stable production process and output. Ideally, cell culture experiments should preferably explore the effects of the concentrations of a wide range of cell culture medium components (such as sugars, metal ions, or amino acids, etc.) on the cell line at every moment, but this is a challenge for the entire biopharmaceutical industry. It not only requires a large amount of time, manpower, and material costs, but also has very high requirements for technical capabilities. The application of the Raman spectroscopy analysis model enables real-time and rapid monitoring of various parameters during the cell culture process, which can shorten the process development time and save test costs during the small-scale laboratory process development stage; and reduce the differences in product quality and output between batches during the large-scale production stage.

[0005] The feeding operation during the cell culture process is a key process action in the entire process, and it has an irreversible and important impact on the cell culture process. Therefore, developing a real-time dynamic automated feeding strategy has important application value. Summary of the Invention

[0006] In view of the deficiencies in the prior art, the purpose of the present invention is to provide an automated feeding method based on real-time dynamic detection results. The present invention adopts process analysis technology to timely feedback the concentration of components in the cell culture medium, and uses the real-time concentration to achieve a dynamic feeding method that can be automatically fed on demand and without human intervention. It solves the complex and inefficient process development strategy obtained by traditional large-scale feeding parallel experiments and artificial stage sampling methods during the process of cell culture for producing recombinant proteins.

[0007] To achieve the object of this invention, the present invention adopts the following technical solutions:

[0008] In a first aspect, the present invention provides an automated feeding method based on real-time dynamic detection results, and the automated feeding method includes:

[0009] (1) Collect real-time dynamic detection results, calculate the consumption rate and generation rate of each component per unit time; select the component with the fastest consumption rate as the marker for the current time period;

[0010] (2) Collect historical detection data, calculate the cell growth rate in the logarithmic growth phase and the target product expression rate in the culture harvest stage; use the concentration level of the marker when the cell growth rate parameter in the logarithmic growth phase is the largest as the component concentration setting value for the current time period in the logarithmic growth phase; use the concentration level of the marker when the target product expression rate parameter in the culture harvest stage is the largest as the component concentration setting value for the current time period in the culture harvest stage;

[0011] (3) Calculate the marker addition dose in the current time period according to the consumption rate and generation rate of the marker, calculate the addition dose of the culture medium according to the marker addition dose, and obtain the feeding strategy.

[0012] Cell culture has now been widely used in the research and development and production of biological drugs. It is extremely crucial to maintain an ideal level of nutrients at each stage of its culture process, because it greatly affects the yield, quality and input cost of biological drugs. In order to ensure that cells grow in the optimal environment, it is extremely important to optimize the feeding control method. The traditional feeding method is to set different ratios to feed daily or every other day according to the experience of the culture process. This method not only requires a lot of manpower to sample, measure samples, and calculate the required feeding amount according to the off-line monitoring data, but also requires equipment (balance, peristaltic pump). Moreover, this feeding method cannot be monitored in a timely manner, which may cause nutrient deficiency or excess at a certain moment during the cell culture process.

[0013] In the present invention, according to the time increment data, the consumption rate and generation rate of components in different time periods are calculated, and markers are selected in real time dynamically; according to the time increment data, the consumption rate and generation rate of components in different time periods are calculated, and the feeding amount is calculated in real time dynamically. The present invention controls the feeding operation by the nutrient concentration in the cell culture medium, avoiding a large number of samplings to reduce manual operations and avoiding human errors. At the same time, through real-time monitoring and automatic control, it is ensured that the nutrient level in the cell culture medium remains at a reasonable level that meets the actual needs of the cells throughout the culture process.

[0014] Preferably, in step (1), the screening step of the marker includes:

[0015] Collect the current real-time dynamic detection results, and calculate the consumption rate and generation rate of each component per unit time currently;

[0016] Collect the historical real-time dynamic detection results, and calculate the consumption rate and generation rate of each component per unit time historically;

[0017] Combine the consumption rate and generation rate of each component per unit time currently with the consumption rate and generation rate of each component per unit time historically, and screen out the component with the fastest consumption rate as the marker for the current time period.

[0018] Preferably, in step (2), the historical detection data is the historical detection data of the fermentation process with the same cell culture process mode as this time.

[0019] Preferably, in step (3), the step of calculating the addition dose of the culture medium includes:

[0020] m 补 =(G 设 -G 实 )×V 培 / G 补

[0021] wherein, m 补 represents the amount of the feeding culture medium to be added, G 设 represents the set marker concentration, G 实 represents the current marker concentration detected by the model real-time analysis Raman spectrum, V 培 represents the volume of the cell culture medium in the bioreactor, G 补 represents the marker concentration in the feeding culture medium.

[0022] Preferably, in step (3), the feeding culture medium includes a specific proportion of the marker and other components, and the addition amounts of other components are obtained according to the addition dose of the marker.

[0023] Preferably, in step (3), after obtaining the feeding strategy, it further includes writing the feeding strategy into the control system.

[0024] Preferably, after the control system receives the numerical instructions of the feeding amount, set value, and feeding time point, it associates with the peristaltic pump hardware and feeds instantaneously in a dosage manner.

[0025] In the present invention, the flowchart of the automatic feeding method is as Figure 1 shown. First, the output parameters of the process analysis technology are digitized and entered into the software, the culture time is matched, and the consumption and generation rates of each component per unit time are calculated. According to the time increment data, the consumption and generation rates of different components in different time periods are calculated. The concentration of the component with the fastest consumption rate per unit time is judged and selected. This component is selected as the marker for the current time period. According to the cell culture process mode and data, the cell growth rate in the logarithmic growth phase and the target protein expression rate in the culture harvest stage are calculated respectively. When these two parameters are the largest, the concentration levels of each component are recorded respectively, and the concentration level of this component is selected as the set value for the current time period. The marker addition dosage of the marker in the current time period is calculated through the consumption and generation rates of the marker, and at the same time, the amounts of other nutrient components added to the bioreactor are calculated according to the concentration ratios of each component in the culture medium. The numerical values of all feeding strategies are written into the control system through OPC to complete the operation.

[0026] In a second aspect, the present invention provides an automatic feeding model based on real-time dynamic detection results, and the automatic feeding model includes:

[0027] Marker screening unit: used to collect real-time dynamic detection results, calculate the consumption rate and generation rate of each component per unit time; select the component with the fastest consumption rate as the marker for the current time period;

[0028] Marker concentration setting unit: used to collect historical detection data, calculate the cell growth rate in the logarithmic growth phase and the target product expression rate in the culture harvest stage; take the concentration level of the marker when the cell growth rate parameter in the logarithmic growth phase is the largest as the component concentration set value for the current time period in the logarithmic growth phase; take the concentration level of the marker when the target product expression rate parameter in the culture harvest stage is the largest as the component concentration set value for the current time period in the culture harvest stage;

[0029] Feeding amount calculation unit: used to calculate the marker addition dosage in the current time period according to the consumption rate and generation rate of the marker, calculate the addition dosage of the culture medium according to the marker addition dosage, and obtain the feeding strategy.

[0030] Preferably, in the marker screening unit, the screening steps of the marker include:

[0031] Collect the current real-time dynamic detection results, and calculate the consumption rate and generation rate of each component within the current unit time;

[0032] Collect the historical real-time dynamic detection results, and calculate the consumption rate and generation rate of each component within the historical unit time;

[0033] Merge the consumption rate and generation rate of each component within the current unit time with the consumption rate and generation rate of each component within the historical unit time, and screen out the component with the fastest consumption rate as the marker for the current time period.

[0034] Preferably, in the marker concentration setting unit, the historical detection data is the historical detection data of the fermentation process with the same cell culture process mode as the current one.

[0035] Preferably, in the feeding amount calculation unit, the steps of calculating the addition dose of the culture medium include:

[0036] m 补 =(G 设 -G 实 )×V 培 / G 补

[0037] Wherein, m 补 represents the amount of the feeding culture medium to be added, G 设 represents the set marker concentration, G 实 represents the current marker concentration detected by the model's real-time analysis of Raman spectroscopy, V 培 represents the volume of the cell culture solution in the bioreactor, and G 补 represents the marker concentration in the feeding culture medium.

[0038] Preferably, in the feeding amount calculation unit, the feeding culture medium includes a specific proportion of the marker and other components, and the addition amounts of other components are obtained according to the addition dose of the marker.

[0039] Preferably, in the feeding amount calculation unit, after obtaining the feeding strategy, it further includes writing the feeding strategy into the control system.

[0040] Preferably, after receiving the numerical instructions of the feeding amount, set value, and feeding time point, the control system associates with the peristaltic pump hardware and performs instantaneous feeding in a dosage manner.

[0041] In a third aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored. Wherein, when the computer program is executed by a processor, the steps of the automated feeding method based on real-time dynamic detection results described in the first aspect are implemented.

[0042] Fourthly, the present invention provides a computer device, including a memory and a processor, and a computer program capable of running on the processor is stored on the memory. When the processor executes the computer program, the steps of the automated feeding method based on the real-time dynamic detection result described in the first aspect are implemented.

[0043] The present invention also provides a cell culture feeding control device, which can realize real-time cell culture feeding control. Each unit in the device implements the following operation steps, including:

[0044] (1) Using the component concentration of the cell culture medium fed back by the process analysis tool as the feedback control marker.

[0045] 1.1 The process analysis tool includes, but is not limited to, Raman spectrometer, infrared spectrometer, on-line biochemical analyzer, on-line high performance liquid chromatograph, on-line mass spectrometer.

[0046] 1.2 Use the method of Example 1 to select the feedback control marker.

[0047] (2) Map the component concentration value to the process value position of the controller control module through the data transmission protocol (OPC / moudbus).

[0048] 2.1 The controller includes, but is not limited to, Finesse, GE or STR.

[0049] (3) Load the multivariate analysis processing component concentration logic into the control module to calculate the feeding amount. Identify the process characteristics to determine the feeding time.

[0050] 3.1 The multivariate analysis processing control module is developed based on the following software, including but not limited to SIMCA, Rstudio, Matlab.

[0051] (4) Set the correlation formula between the feeding amount and the feeding time.

[0052] (5) Complete all action operations based on the feeding amount and the feeding time.

[0053] (6) Or calculate the amount of nutrients required by the cells at this moment when a new component concentration value is input each time, and map the calculated data value to the peristaltic pump of the controller.

[0054] (7) Complete all action operations, complete the fermentation control process, and output the parameters of the fermentation process control.

[0055] The numerical ranges described in the present invention not only include the point values listed above, but also any point values between the above numerical ranges that are not listed. Due to space limitations and for the sake of brevity, the present invention does not exhaustively list the specific point values included in the said ranges.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] The method of the present invention can improve the monitoring and control during the cell culture process, respond to the nutritional requirements of the dynamic changes of cells; can adjust the addition of the mixed feeding medium according to the concentrations of the components in the cell culture environment; provide the optimal culture environment for cells, stabilize the process to the greatest extent and increase the yield; can generate feeding strategies intelligently to improve efficiency; perform automated feeding operations according to the generated feeding strategies, reduce human errors and reduce the risk of contamination. The said method improves the speed and accuracy of cell culture process development; can provide customized feeding media for different cell lines, clones and processes; can be quickly applied to the process development process and production processes of different scales. Description of the Drawings

[0058] Figure 1 is a flowchart of the automated feeding method.

[0059] Figure 2 is a flowchart of the feedback control application.

[0060] Figure 3 is a real-time dynamic feeding diagram (I).

[0061] Figure 4 is a real-time dynamic feeding diagram (II).

[0062] Figure 5 is an effect diagram of the comparison of viable cell density.

[0063] Figure 6 is an effect diagram of the comparison of cell viability.

[0064] Figure 7 is an effect diagram of the comparison of glucose concentration.

[0065] Figure 8 is an effect diagram of the comparison of lactate concentration.

[0066] Figure 9 is an effect diagram of the comparison of lactate dehydrogenase concentration.

[0067] Figure 10 is an effect diagram of the comparison of the expression level of the target protein.

[0068] Figure 11 is an effect diagram of the comparison of the protein expression level of a single cell.

[0069] Figure 12It is the comparison effect diagram of ammonia ion concentration.

[0070] Figure 13 It is the comparison effect diagram (I) of amino acid concentration between manual feeding and dynamic feeding when the marker is glucose.

[0071] Figure 14 It is the comparison effect diagram (II) of amino acid concentration between manual feeding and dynamic feeding when the marker is glucose.

[0072] Figure 15 It is the comparison effect diagram of viable cell density.

[0073] Figure 16 It is the comparison effect diagram of cell viability.

[0074] Figure 17 It is the comparison effect diagram of glucose concentration.

[0075] Figure 18 It is the comparison effect diagram of lactic acid concentration.

[0076] Figure 19 It is the comparison effect diagram of target protein concentration.

[0077] Figure 20 It is the schematic diagram of the cell culture feeding control device. Detailed implementation mode

[0078] The technical solution of the present invention will be further described below through specific implementation modes. Those skilled in the art should understand that the described embodiments are only to help understand the present invention and should not be regarded as specific limitations on the present invention.

[0079] For those not specifying specific techniques or conditions in the embodiments, they shall be carried out according to the techniques or conditions described in the literature in this field or according to the product specifications. For reagents or instruments not specifying the manufacturer, they are all conventional products that can be obtained through regular commercial channels.

[0080] Comparative example 1

[0081] This comparative example adopts the traditional manual timed feeding strategy. Traditional feeding is classified according to the amount of feeding or the physical and chemical properties of the feeding. Usually, the feeding is divided into three categories: "large feeding" with weakly acidic feeding and a large amount of feeding, "small feeding" with alkaline feeding and a small amount of feeding, and separate glucose feeding. The feeding amount and feeding time of large and small feedings are determined by statistical analysis after obtaining the rate of cell consumption of nutrients through cell culture experiments.

[0082] The manual timed feeding strategy in this comparative example is to feed every other day, that is, add the feeding medium to the reactor manually on the 2nd, 4th, 6th, 8th, 10th, and 12th days of the culture time. The calculation formula for the feeding amount is:

[0083] On the 2nd, 4th, and 6th days, m 大补 = V 培 × 3%; m 小补 = V 培 × 0.3%

[0084] On the 8th and 10th days, m 大补 = V 培 × 4%; m 小补 = V 培 × 0.4%

[0085] On the 12th day, m 大补 = V 培 × 1%; m 小补 = V 培 × 0.1%

[0086] The strategy for glucose supplementation is generally to control the glucose concentration within an appropriate range based on experience or experimental data. According to the off-line biochemical detection equipment, the residual glucose concentration in the cell culture medium is detected, and the amount of glucose to be supplemented every day is calculated. The formula for glucose consumption rate is as follows:

[0087] C 葡萄糖 = (R 葡萄糖 + F 葡萄糖 ) / T

[0088] Wherein, C 葡萄糖 represents the glucose consumption rate, R 葡萄糖 represents the residual glucose concentration in the cell culture medium, F 葡萄糖 represents the glucose concentration in the feed, and T represents the time interval.

[0089] Based on the above formula, the glucose consumption rate is calculated, and then the formula for the glucose supplementation strategy is formulated according to the consumption rate. Based on the calculation results, an appropriate amount of glucose is supplemented to keep the glucose within a certain level range.

[0090] The formula for glucose supplementation in manual timed feeding operation is as follows:

[0091] m 补 = (G 设 - G 实 ) × V 培 / G 补

[0092] Wherein, m 补 represents the amount of the feed medium to be added, G 设 represents the set glucose concentration, G 实 represents the current glucose concentration detected by the model's real-time analysis of Raman spectroscopy, V 培 represents the volume of the cell culture medium in the bioreactor, and G 补 represents the glucose concentration in the feed medium.

[0093] Example 1

[0094] The automated feeding method based on real-time dynamic detection results is used to control the automated feeding of Case 1. At the same time, a comparative experiment is carried out by adopting the traditional manual timed feeding strategy in Comparative Example 1.

[0095] In this embodiment, the flowchart of feedback control is as Figure 2 shown.

[0096] The feedback control flowchart covers a variety of cell culture processes, such as fed-batch culture process (batch feeding); perfusion process (continuous feeding). In both processes, process analysis technology can be used to monitor the component concentration in the cell culture medium. After analyzing and calculating each concentration, the method of this patent is used to control each process operation control point in the figure. Including but not limited to, batch feeding operation point, continuous feeding operation point, cell discharging operation point (to stabilize the perfusion process, a certain amount of cell fluid needs to be discharged from the reactor in real time or regularly), pH control operation point (adding alkali), ventilation control operation, harvest rate control operation.

[0097] The experimental materials and related experimental methods in Case 1 are as follows:

[0098] In this embodiment, a cell line of CHO-K1 producing monoclonal antibody is used in the cell culture test part. The culture conditions of this cell line in the seed stage in a shaker (Kuhner) are 36.5 °C, a rotation speed of 110 rpm, a CO2 concentration level of 6%, and the culture medium used is a self-developed culture medium (MagniCHO). In the production culture stage, a 3L reactor is used, with an initial culture volume of 1.5L, a culture temperature of 36.5 °C, a set pH value of 6.90 ± 0.25, a dissolved oxygen saturation of 40%, and an initial inoculation density of 1.0×10 6 cells / mL. The feeding medium is also a self-developed culture medium (MagniCHO Feed1). The materials are shown in Table 1.

[0099] Table 1

[0100] Material Name Article Number Manufacturer MagniCHO CR80262 Merck MagniCHO Feed 1 CR80270 Merck

[0101] The high-dimensional data analysis and logical calculation of the process are realized by using the open-source programming tool R (version 4.0.5).

[0102] The automated feeding method based on real-time dynamic detection results includes the following steps:

[0103] (1) Collect the current real-time dynamic detection results, calculate the consumption rate and generation rate of each component within the current unit time; collect the historical real-time dynamic detection results, calculate the consumption rate and generation rate of each component within the historical unit time; merge the consumption rate and generation rate of each component within the current unit time with those within the historical unit time, and screen out the component with the fastest consumption rate as the marker for the current time period.

[0104] (2) Use the historical detection data of the fermentation process with the same cell culture process mode as this time, calculate the cell growth rate during the logarithmic growth phase and the target product expression rate during the culture harvest phase, and statistically analyze the concentration levels of each component when these two parameters are maximized. Use the concentration level corresponding to the selected marker as the set value of the component concentration for the current time period.

[0105] (3) Calculate the addition dose of the marker during the current time period based on the consumption rate and generation rate of the marker, calculate the addition dose of the culture medium based on the addition dose of the marker, and obtain the feeding strategy. Write the feeding strategy into the control system; after the control system receives the numerical instructions of the feeding amount, set value, and feeding time point, it associates with the peristaltic pump hardware and performs instantaneous feeding in a dosage manner. Stop the control operation after completing the current operation until the next feeding instruction is issued.

[0106] The steps for calculating the addition dose of the culture medium include:

[0107] m 补 =(G 设 -G 实 )×V 培 / G 补

[0108] where m 补 represents the amount of the feeding culture medium to be added, G 设 represents the set marker concentration, G 实 represents the current marker concentration detected by the model's real-time analysis of Raman spectroscopy, V 培 represents the volume of the cell culture solution in the bioreactor, and G 补 represents the marker concentration in the feeding culture medium.

[0109] Summarize the feeding strategy for each time to obtain the automated feeding strategy for this fermentation. Select glucose as the marker for fermentation control. The feeding time and feeding amount of glucose are as shown in Figure 3 and Figure 4 . Figure 3 is the real-time dynamic feeding diagram (one), Figure 4 is the real-time dynamic feeding diagram (two).

[0110] It can be found from the operation schematic diagrams of manual timed feeding and real-time dynamic feeding that during the manual timed feeding process, a total of 7 feeding operations occurred (indicated by the blue line ends below), and the corresponding marker (glucose) also had a concentration change instantaneously during feeding (indicated by the yellow line segments above). For real-time dynamic feeding, the feeding amount is evenly distributed throughout the cell culture process, and the marker always remains near the set value.

[0111] Fermentation control is carried out according to the strategy of automated feeding. In addition, other unselected components are used as markers for fermentation control to conduct comparative experiments. The unselected components include: aspartic acid, total asparagine, histidine, methionine current combination.

[0112] The fermentation results of the above-mentioned strategy of automated feeding are as follows:

[0113] Figure 5 For the comparison effect diagram of viable cell density, the results show that the manual timed feeding action and the automatic dynamic feeding action do not significantly affect the viable cell density during the cell culture process. The growth trends and the highest density values of the viable cell density parameters under the two feeding actions are comparable.

[0114] Figure 6 For the comparison effect diagram of cell viability, the results show that although manual timed feeding can control the glucose level, due to the long lag time and fluctuating glucose concentration, the lactic acid concentration remains high, so the cell viability is low.

[0115] Figure 7 For the comparison effect diagram of glucose concentration, the results show that manual timed feeding regulates the feeding and sugar supplementation amounts for the next day or the third day according to the data detected offline every day. This method has serious lag and there are large fluctuations in the glucose concentration in the cell culture medium.

[0116] Figure 8 For the comparison effect diagram of lactic acid concentration, the results show that the traditional method controls lactic acid by regulating the feeding and sugar supplementation amounts for the next day or the third day according to the data detected offline every day. Manual timed feeding has serious lag and there are large fluctuations in the glucose concentration in the cell culture medium. Therefore, it is difficult to control the lactic acid level by controlling the glucose concentration, resulting in a still high lactic acid concentration.

[0117] Figure 9 For the comparison effect diagram of lactate dehydrogenase concentration, the results show that the concentration of lactate dehydrogenase in the traditional manual feeding operation is higher than that in the real-time dynamic feeding operation. Due to the higher lactic acid concentration, the pH value of the cell culture medium is higher, and the rapid decline of cell viability further deteriorates the state of the culture medium. More cells die and rupture, releasing more lactate dehydrogenase.

[0118] Figure 10 It is the comparison effect diagram of the target protein expression level, where Figure 11 is the protein expression level of a single cell. Both of the two figures show that the yields of real-time dynamic feeding operation and manual timed feeding operation are comparable, and all dynamic feedings are higher than manual feeding. Among them, the dynamic feedings with methionine and glucose as markers are significantly better than manual feeding.

[0119] Figure 12 It is the comparison effect diagram of ammonia ion concentration. The results show that the dynamic feeding operation also controls the level of this metabolic waste, ammonia ion, lower than that of manual feeding. This performance can also improve the viable cell density and cell viability, thereby ensuring high yield of the target protein.

[0120] Figure 13 It is the comparison effect diagram (I) of amino acid concentrations between manual feeding and dynamic feeding when the marker is glucose. Figure 14 It is the comparison effect diagram (II) of amino acid concentrations between manual feeding and dynamic feeding when the marker is glucose. The results show that the concentration levels of different amino acid types are different, and the concentration change range of dynamic feeding is more stable and has a smaller change amplitude than that of manual feeding. This may bring a more stable nutritional environment, which is more suitable for cell growth and protein synthesis. Figure 13 The level of 5 - 6 mM is proline; the level of 2 - 4 mM is leucine; the level of 0 - 2 mM is cysteine. Figure 14 The level of 3 - 6 mM is threonine; the level of 2 - 4 mM is lysine; the level of 0 - 1 mM is tryptophan. Timed manual feeding (square marker); real-time dynamic feeding (circular marker).

[0121] In summary, in Case 1, through calculation and screening, glucose was finally used as the marker. And in Case 1, there was a problem of lactic acid accumulation during the cultivation process. Selecting glucose as the marker not only serves to achieve subsequent automatic feeding control, but also can control the lactic acid level by controlling the glucose concentration. The traditional method is to regulate the feeding and sugar supplementation amounts for the next day or the third day based on the data detected offline every day to control lactic acid. This method has serious lag, and there is a large fluctuation range in the glucose concentration in the cell culture medium (such as Figure 7 the circular marker curve). Although the glucose level is controlled, due to the long-term lag and fluctuating glucose concentration, the lactic acid concentration is still relatively high, so the cell density and viability are low. The real-time dynamic feeding action in this embodiment effectively controls the glucose concentration within a lower range throughout the cultivation process (such as Figure 7 the diamond marker curve), and at the same time maintains the lactic acid level within a low concentration range at the end of cell culture ( Figure 8 diamond marker curve), thereby maintaining a higher cell viability ( Figure 6The diamond-shaped marker curve), and finally at the end of cell culture, cells can synthesize more target proteins more efficiently ( Figure 10 ). At the same time, the lactate dehydrogenase concentration curve also corroborates the phenomenon of low viability at the end of culture ( Figure 9 ). Figure 13 , Figure 14 The trend in the amino acid comparison effect diagram of can also prove that when glucose is used as the feeding marker, real-time dynamic feeding can reduce the fluctuation of amino acid concentration in the cell culture medium and keep the amino acid concentration within a stable range all the time.

[0122] In the comparative experiment, aspartate and asparagine sum or histidine or methionine were selected as markers. During cell culture, asparagine and aspartate will achieve dynamic balance through acyl transfer. When selecting markers, choosing only one of them will cause errors in feeding operations, resulting in cell culture failure. Therefore, the sum of the concentrations of the two substances in the cell culture medium is used as a marker to control the feeding operation. Through Figure 12 It can be found that another toxic by-product ammonium ion in the cell culture medium has also reduced the ammonium ion concentration in the cell culture medium by 100% due to the real-time dynamic feeding operation using amino acid concentration as a feedback control marker. This phenomenon also helps to improve cell viability and cell state, and helps to increase the yield of the target protein. At the same time, it can be found through the curve controlled by amino acids that the performance of other process parameters is similar.

[0123] Example 2

[0124] This example examines the influence of different ranges of glucose concentration on fermentation based on the automated feeding strategy in Example 1, and further verifies the advantages of the automated feeding strategy of the present invention.

[0125] Using glucose as a marker, the same real-time dynamic feeding action was adopted, and feeding was carried out at high concentration levels and low concentration levels respectively. The fermentation test results are as follows:

[0126] Figure 15 is the comparison effect diagram of viable cell density. The results show that a higher glucose concentration level can maintain a higher biomass (viable cell density level).

[0127] Figure 16 is the comparison effect diagram of cell viability. The results show that, however, a higher glucose level is relatively poor for maintaining viability in the later stage of culture, which may be because a higher glucose concentration causes more lactic acid production.

[0128] Figure 17 is the comparison effect diagram of glucose concentration. The results show that the glucose concentration of the diamond curve is higher, and its concentration level is about 5-6 g / L.

[0129] Figure 18 This is the effect comparison diagram of lactic acid concentration. The results show that high-concentration glucose cannot maintain a lower lactic acid level. This may be because the enrichment of glucose in the cell glycolysis pathway causes excessive lactic acid accumulation, while glucose does not enter the tricarboxylic acid cycle pathway.

[0130] Figure 19 This is the effect comparison diagram of the target protein concentration. The results show that too high a lactic acid level causes the pH value of the culture medium to be too low. The state of the cells changes at too low a pH value, and the protein synthesis pathway is blocked, resulting in a lower yield.

[0131] In summary, when glucose is selected as the feeding marker and different ranges of glucose concentrations are given, and the same real-time dynamic feeding action is adopted, it is also proved that a lower level of glucose concentration can effectively control the lactic acid concentration, thereby ensuring cell viability and a stable yield. The above comparative experiments show that the automated feeding strategy of the present invention can automatically generate a feeding strategy according to the real-time detection results, and the fermentation effect of the feeding strategy is better than that of the traditional feeding strategy.

[0132] Example 3

[0133] A cell culture feeding control device includes:

[0134] (1) Using the component concentration of the cell culture medium feedback by the process analysis tool as the feedback control marker.

[0135] 1.1 The process analysis tool includes, but is not limited to, Raman spectrometer, infrared spectrometer, on-line biochemical analyzer, on-line high performance liquid chromatograph, on-line mass spectrometer.

[0136] 1.2 Use the method of Example 1 to select the feedback control marker.

[0137] (2) Map the component concentration value to the process value position of the controller control module through the data transmission protocol (OPC / moudbus).

[0138] 2.1 The controller includes, but is not limited to, Finesse, GE or STR.

[0139] (3) Load the multi-variable analysis processing component concentration logic into the control module to calculate the feeding amount. Identify the process characteristics to determine the feeding time.

[0140] 3.1 The multi-variable analysis processing control module is developed based on the following software, including but not limited to SIMCA, Rstudio, Matlab.

[0141] (4) Set the correlation formula for the feeding amount and the feeding time.

[0142] (5) Complete all action operations based on the feeding volume and feeding time.

[0143] (6) Or calculate the amount of nutrients required by the cells at that moment when a new component concentration value is input each time, and map the calculated data value to the peristaltic pump of the controller.

[0144] (7) Complete all action operations. For the specific effects, see Figures 4 - 19 .

[0145] Figure 20 It is a schematic diagram of the feeding control device for cell culture.

[0146] In summary, the present invention controls the feeding operation by the nutrient concentration in the cell culture solution, avoiding a large amount of sampling and reducing personnel operations, and avoiding human errors. At the same time, through real-time monitoring and automatic control, it is ensured that the nutrient level in the cell culture solution remains at a reasonable level of the actual needs of the cells throughout the culture process.

[0147] The applicant declares that the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention fall within the protection scope and the public disclosure scope of the present invention.

Claims

1. An automated feeding method based on real-time dynamic detection results, characterized in that, The automated feeding method includes: (1) Collect real-time dynamic detection results, calculate the consumption rate and generation rate of each component per unit time; select the component with the fastest consumption rate as the marker for the current time period; (2) Collect historical detection data, calculate the cell growth rate in the logarithmic growth phase and the target product expression rate in the culture harvest phase; use the concentration level of the marker when the cell growth rate parameter in the logarithmic growth phase is the largest as the component concentration setting value for the current time period in the logarithmic growth phase; use the concentration level of the marker when the target product expression rate parameter in the culture harvest phase is the largest as the component concentration setting value for the current time period in the culture harvest phase; (3) Calculate the marker addition dose in the current time period according to the consumption rate and generation rate of the marker, calculate the addition dose of the culture medium according to the marker addition dose, and obtain the feeding strategy.

2. The automated feeding method based on real-time dynamic detection results according to claim 1, wherein In step (1), the screening steps of the marker include: Collect the current real-time dynamic detection results, and calculate the consumption rate and generation rate of each component per unit time currently; Collect the historical real-time dynamic detection results, and calculate the consumption rate and generation rate of each component per unit time historically; Combine the consumption rate and generation rate of each component per unit time currently with the consumption rate and generation rate of each component per unit time historically, and screen out the component with the fastest consumption rate as the marker for the current time period.

3. The automated feeding method based on real-time dynamic detection results according to claim 1 or 2, characterized in that In step (2), the historical detection data is the historical detection data of the fermentation process with the same cell culture process mode as this time.

4. The automated feeding method based on real-time dynamic detection results according to any one of claims 1-3, characterized in that In step (3), the steps of calculating the addition dose of the culture medium include: m 补 = (G 设 - G 实 ) × V 培 / G 补 Among them, m 补 represents the amount of the feeding medium to be added, G 设 represents the set marker concentration, G 实 represents the current marker concentration detected by the model for real-time analysis of Raman spectroscopy, V 培 represents the volume of the cell culture medium in the bioreactor, G 补 represents the marker concentration in the feeding medium.

5. The automated feeding method based on real-time dynamic detection results according to any one of claims 1-4, characterized in that In step (3), the feeding culture medium includes a specific proportion of the marker and other components, and the addition amount of other components is obtained according to the addition dose of the marker.

6. The automated feeding method based on real-time dynamic detection results according to any one of claims 1-5, characterized in that In step (3), after obtaining the feeding strategy, it further includes writing the feeding strategy into the control system; Preferably, after the control system receives the numerical instructions of the feeding amount, setting value, and feeding time point, it associates with the peristaltic pump hardware and feeds instantaneously in a dosage manner.

7. An automated feeding model based on real-time dynamic detection results, characterized in that, The automated feeding model includes: Marker screening unit: used to collect real-time dynamic detection results, calculate the consumption rate and generation rate of each component per unit time; select the component with the fastest consumption rate as the marker for the current time period; Marker concentration setting unit: used to collect historical detection data, calculate the cell growth rate in the logarithmic growth phase and the target product expression rate in the culture harvest phase; use the concentration level of the marker when the cell growth rate parameter in the logarithmic growth phase is the largest as the component concentration setting value for the current time period in the logarithmic growth phase; use the concentration level of the marker when the target product expression rate parameter in the culture harvest phase is the largest as the component concentration setting value for the current time period in the culture harvest phase; Feeding amount calculation unit: used to calculate the marker addition dose in the current time period according to the consumption rate and generation rate of the marker, calculate the addition dose of the culture medium according to the marker addition dose, and obtain the feeding strategy.

8. The automated feeding model based on real-time dynamic detection results according to claim 7, characterized in that In the marker screening unit, the screening steps of the marker include: Collect the current real-time dynamic detection results, and calculate the consumption rate and generation rate of each component per unit time currently; Collect real-time dynamic detection results of history, and calculate the consumption rate and generation rate of each component within a historical unit time; Merge the consumption rate and generation rate of each component within the current unit time with the consumption rate and generation rate of each component within the historical unit time, and screen out the component with the fastest consumption rate as the marker for the current time period; Preferably, in the marker concentration setting unit, the historical detection data is the historical detection data of the fermentation process with the same cell culture process mode as the current one; Preferably, in the feeding amount calculation unit, the step of calculating the addition dose of the culture medium includes: m 补 = (G 设 - G 实 ) × V 培 / G 补 Among them, m 补 represents the amount of the feeding medium to be added, G 设 represents the set marker concentration, G 实 represents the current marker concentration detected by the model's real-time analysis of Raman spectroscopy, V 培 represents the volume of the cell culture medium in the bioreactor, G 补 represents the marker concentration in the feeding medium; Preferably, in the feeding amount calculation unit, the feeding culture medium includes a specific proportion of the marker and other components, and the addition amount of other components is obtained according to the addition dose of the marker; Preferably, in the feeding amount calculation unit, after obtaining the feeding strategy, it further includes writing the feeding strategy into the control system; Preferably, after receiving the numerical instructions of the feeding amount, set value, and feeding time point, the control system associates with the peristaltic pump hardware and feeds instantaneously in a dosage manner.

9. A computer-readable storage medium, characterized in that, A computer program is stored on the storage medium, wherein when the computer program is executed by a processor, the steps of the automated feeding method based on real-time dynamic detection results according to any one of claims 1-6 are implemented.

10. A computer device, comprising a memory and a processor, wherein a computer program capable of running on the processor is stored on the memory, characterized in that When the processor executes the computer program, the steps of the automated feeding method based on real-time dynamic detection results according to any one of claims 1-6 are implemented.

Images