Information processing device, information processing method, and program

Abstract

An information processing device (14): acquires, on the basis of a result from each of a plurality of performances of a cell culture process, a regression curve indicating the relationship between the culture time and a physical quantity correlated with the number of cells; acquires a first parameter including a plurality of parameter elements included in a regression formula that indicates the regression curve; acquires a second parameter including a plurality of parameter elements on the basis of a tangent line at an inflection point (72) of the regression curve; acquires a third parameter including a plurality of parameter elements by performing principal component analysis on the first parameter and the second parameter; and generates correlation information indicating the relationship between two parameter elements among the parameter elements for each of the performances of the cell culture process.

Description

Information Processing Apparatus, Information Processing Method and Program

[0001] The present disclosure relates to an information processing apparatus, an information processing method and a program.

[0002] Japanese Unexamined Patent Publication No. 2024-36914 discloses an importance determination apparatus (and an importance determination method) relating to cell culture. This importance determination apparatus acquires calculation results of the importance of each of one or more types of experimental conditions for each of one or more types of experimental results in an experiment for maturing cells.

[0003] Japanese Unexamined Patent Publication No. 2024-36914

[0004] A technique capable of performing analysis from various perspectives is desired for cell culture.

[0005] An object of the present disclosure is to solve the problem described above.

[0006] (1) One aspect of the present disclosure is an information processing apparatus comprising: a regression curve acquisition unit that acquires, based on a result of cell culture, a regression curve indicating a relationship between a physical quantity correlated with the number of cells and culture time; a first parameter acquisition unit that acquires a first parameter including a plurality of parameter elements contained in a regression equation representing said regression curve; a second parameter acquisition unit that acquires a second parameter including a plurality of parameter elements based on a tangent line at an inflection point of said regression curve, wherein said first parameter acquisition unit acquires said first parameter for each of a plurality of cell cultures performed multiple times, and said second parameter acquisition unit acquires said second parameter for each of a plurality of cell cultures performed multiple times; a third parameter acquisition unit that acquires a third parameter including a plurality of parameter elements by performing principal component analysis on said first parameter and said second parameter; and an information generation unit that generates correlation information, which is information indicating a relationship between one parameter element and another parameter element for each of a plurality of cell cultures performed multiple times.

[0007] According to such a configuration, analysis from a new perspective can be performed regarding cell culture.

[0008] (2) In the information processing device described in item (1) above, each parameter element included in the third parameter may be the score of each principal component generated by the principal component analysis.

[0009] (3) The information processing device described in item (1) or (2) above may be provided with a display control unit that displays the correlation information on a display screen.

[0010] (4) In the information processing device described in item (3) above, the display control unit may display the scatter plot, which is the correlation information, on the display screen.

[0011] (5) In the information processing device described in item (4) above, the display control unit may, in addition to the scatter plot, display on the display screen the contribution of each parameter element of the first parameter and the contribution of each parameter element of the second parameter in each principal component generated by the principal component analysis.

[0012] (6) The information processing device described in any one of the above items (1) to (5) includes a parameter element determination unit that determines which parameter elements contribute to the success or failure of cell culture from among a plurality of parameter elements included in the first parameter and a plurality of parameter elements included in the second parameter, wherein the third parameter for each of the multiple cell cultures performed and the culture success or failure information, which is information indicating the success or failure of cell culture, are associated with each other, and the parameter element determination unit may determine which parameter elements contribute to the success or failure of cell culture based on the contribution of each parameter element of the first parameter and each parameter element of the second parameter in each principal component generated by the principal component analysis, the scatter plot which is the correlation information, and the culture success or failure information.

[0013] (7) Another aspect of the present disclosure is an information processing method comprising: a regression curve acquisition step of acquiring a regression curve showing a relationship between a physical quantity correlated with the number of cells and culture time based on the results of cell culture; a first parameter acquisition step of acquiring a first parameter including a plurality of parameter elements included in a regression equation showing the regression curve; a second parameter acquisition step of acquiring a second parameter including a plurality of parameter elements based on a tangent line at an inflection point of the regression curve, wherein the first parameter acquisition step acquires the first parameter for each of multiple cell cultures performed, the second parameter acquisition step acquires the second parameter for each of multiple cell cultures performed, and a third parameter acquisition step of acquiring a third parameter including a plurality of parameter elements by performing principal component analysis on the first parameter and the second parameter; and an information generation step of generating correlation information which is information showing the relationship between one parameter element and other parameter elements for each of multiple cell cultures performed.

[0014] This configuration allows for analysis of cell culture from a new perspective.

[0015] (8) In the information processing method described in item (7) above, each parameter element included in the third parameter may be the score of each principal component generated by the principal component analysis.

[0016] (9) The information processing method described in item (7) or (8) above may include a display step of displaying the correlation information on a display screen.

[0017] (10) In the information processing method described in item (9) above, the display step may include displaying the scatter plot, which is the correlation information, on the display screen.

[0018] (11) In the information processing method described in item (10) above, in the display step, in addition to the scatter plot, the contribution of each parameter element of the first parameter and the contribution of each parameter element of the second parameter in each principal component generated by the principal component analysis may be displayed on the display screen.

[0019] (12) The information processing method described in any one of the above items (7) to (11) includes a determination step in which a parameter element that contributes to the success or failure of cell culture is determined from among a plurality of parameter elements included in the first parameter and a plurality of parameter elements included in the second parameter, wherein the third parameter for each of the multiple cell cultures performed and the culture success or failure information, which is information indicating the success or failure of cell culture, are associated with each other, and in the determination step, the parameter element that contributes to the success or failure of cell culture may be determined based on the contribution of each parameter element of the first parameter and each parameter element of the second parameter in each principal component generated by the principal component analysis, the scatter plot which is the correlation information, and the culture success or failure information.

[0020] (13) Another aspect of this disclosure is a program for causing a computer to execute the information processing method described in any one of items (7) to (12) above.

[0021] According to this disclosure, it is possible to perform analyses of cell culture from a new perspective. For example, based on information obtained from regression curves showing the relationship between physical quantities correlated with cell number and culture time, it is possible to quantitatively determine the success or failure of the culture and identify parameter elements that contribute to the success or failure of the culture, and it becomes possible to analyze multiple culture data based on the same criteria.

[0022] Figure 1 is a block diagram of the cell culture system. Figure 2 is a diagram showing an example of a decision tree model. Figure 3 is a flowchart of the analysis process performed by the information processing device. Figure 4 is a graph showing the relationship between culture time and lactate production rate. Figures 5A to 5D are graphs showing the contribution (coefficient) of the first and second parameters to each principal component. Figure 6 is a scatter plot showing the relationship between the first principal component (PCA0) and the second principal component (PCA1). Figure 7 is a scatter plot showing the relationship between the first principal component (PCA0) and the third principal component (PCA2). Figure 8 is a scatter plot showing the relationship between the first principal component (PCA0) and the fourth principal component (PCA3). Figure 9 is a scatter plot showing the relationship between the second principal component (PCA1) and the third principal component (PCA2). Figure 10 is a scatter plot showing the relationship between the second principal component (PCA1) and the fourth principal component (PCA3). Figure 11 is a scatter plot showing the relationship between the third principal component (PCA2) and the fourth principal component (PCA3). Figure 12 is a flowchart of the support processing performed by the information processing device. Figure 13 is a scatter plot showing the relationship between parameter a and capA. Figure 14 is a table showing the flow rate of the culture medium. Figure 15 is a graph showing the relationship between culture time and glucose concentration and the relationship between culture time and lactate concentration. Figure 16 is a graph showing the relationship between culture time and the metabolic rate of cells (glucose consumption rate and lactate production rate). Figure 17 is a table showing the number of seeded cells, the number of harvested cells, doubling time, doubling number, and doubling cycle. Figure 18 is a table showing the flow rate of the culture medium. Figure 19 is a graph showing the relationship between culture time and glucose concentration and the relationship between culture time and lactate concentration. Figure 20 is a graph showing the relationship between culture time and the metabolic rate of cells (glucose consumption rate and lactate production rate). Figure 21 is a table showing the number of seeded cells, the number of harvested cells, doubling time, doubling number, and doubling cycle.

[0023] To prevent a decline in the quality of cultured cells, it is advisable to use a large amount of culture medium to provide sufficient nutrients and remove metabolic waste products such as lactic acid and ammonia. However, culture medium is expensive. Therefore, it is preferable to maintain a culture environment above the standard while minimizing the amount of culture medium used. In other words, it is preferable to culture cells under optimal conditions that balance the maintenance of a suitable culture environment with the reduction of culture medium usage.

[0024] For example, the optimal conditions for cell culture are determined by conducting multiple cell culture experiments. However, cell culture experiments require a significant amount of time and high costs. For these reasons, it is difficult to conduct enough experiments to determine the optimal conditions for cell culture. Therefore, the optimal conditions for cell culture have not been established. As a result, currently, users need to determine the cell culture conditions themselves. These conditions vary depending on the user's experience. Consequently, there is a risk of variations in cell quality and cost.

[0025] Therefore, efforts are underway to determine the optimal conditions for future cell cultures by collecting and analyzing information on cell cultures that have already been performed. This is expected to reduce variability in cell quality and cost. The disclosure described below relates to a cell culture system equipped with an information processing device that contributes to the analysis of information related to cell culture (also called cell culture information).

[0026] [1. Configuration of Cell Culture System 10] Figure 1 is a block diagram of the cell culture system 10. The cell culture system 10 comprises a cell culture device 12, an information processing device 14, and a server device 16. The cell culture device 12 and the information processing device 14 are connected to each other via a communication line 18 such as a LAN (Local Area Network) or the Internet. Similarly, the cell culture device 12 and the server device 16 are connected to each other via a communication line 18. Similarly, the information processing device 14 and the server device 16 are connected to each other via a communication line 18. Note that the cell culture system 10 does not necessarily have to include a server device 16.

[0027] The cells cultured in the cell culture device 12 may be mammalian cells or cells derived from mammals. Furthermore, the cells cultured in the cell culture device 12 may be adherent cells or suspension cells. Specifically, examples of cells include HEK293 cells (human fetal kidney cells), ES cells (embryonic stem cells), iPS cells (induced pluripotent stem cells), mesenchymal stem cells, fibroblasts, endothelial cells, and neural stem cells. Examples of suspension cells include Jurkat cells (human cell-derived leukemia cells), T cells, regulatory T cells, tumor-infiltrating lymphocytes, CAR-T cells, and CD34-positive cells. The cells cultured in the cell culture device 12 are not limited to those described above.

[0028] [1-1 Cell Culture Apparatus 12] The cell culture apparatus 12 comprises a cell culture circuit 22 and a control device 24. The control device 24 comprises a processor and memory (neither of which are shown). The cell culture circuit 22 comprises a bioreactor 26 and a plurality of pumps 28. The bioreactor 26 comprises a plurality of hollow fibers. The cell culture circuit 22 is composed of a first circulation circuit including the inside of the hollow fibers provided in the bioreactor 26 and a second circulation circuit including the outside of the hollow fibers provided in the bioreactor 26. In this specification, the portion of the bioreactor 26 included in the first circulation circuit is also called IC (Intracapillary), and the portion of the bioreactor 26 included in the second circulation circuit is also called EC (Extracapillary). A first culture medium containing nutrients is supplied to the first circulation circuit. A second culture medium that does not contain nutrients is supplied to the second circulation circuit. One or more pumps 28 are provided in each of the first and second circulation circuits. When the control device 24 operates one or more pumps 28 arranged in the first circulation circuit, the first culture medium circulates through the first circulation circuit. When the control device 24 operates one or more pumps 28 arranged in the second circulation circuit, the second culture medium circulates through the second circulation circuit.

[0029] A sampling unit 30 is connected to at least one of the first circulation circuit and the second circulation circuit. The sampling unit 30 is equipped with a sampling port for sampling the culture medium. A sensor 32 provided in the sampling unit 30 measures physical quantities (glucose concentration, lactate concentration, etc.) that correlate with the number of cells in the culture medium. The sensor 32 transmits information indicating the measurement result (also called sampling information) to the control device 24. The memory of the control device 24 stores the sampling information. The control device 24 also transmits the sampling information to at least one of the server device 16 and the information processing device 14 via the communication line 18.

[0030] The user can also perform the sampling of the culture medium. In this case, the user samples the culture medium from the sampling port and uses the sensor 32 to measure a physical quantity correlated with the number of cells in the culture medium. The sensor 32 may be provided separately from the cell culture apparatus 12. In this case, the user stores the sampling information in the memory of the control device 24 of the cell culture apparatus 12, the storage unit 42 of the information processing device 14, the memory of the server device 16, etc.

[0031] [1-2 Information Processing Device 14] The information processing device 14 can assist a user in analyzing cell culture information. Alternatively, the information processing device 14 can analyze cell culture information on behalf of the user. For example, the information processing device 14 can determine whether a single cell culture that has already been performed was successful or unsuccessful. The information processing device 14 can also provide the user with information useful for analyzing cell culture information. Furthermore, the information processing device 14 can determine information that contributes to the success or failure of cell culture.

[0032] The information processing device 14 comprises an input unit 36, a communication unit 38, an arithmetic unit 40, a storage unit 42, and a display unit 44. The information processing device 14 may be composed of, for example, a computer (personal computer, tablet terminal, smartphone, etc.).

[0033] The input unit 36 ​​may consist of a human-machine interface operated by the user. The input unit 36 ​​may be, for example, a keyboard, a touch panel, a mouse, etc. The input unit 36 ​​sends the information entered by the user to the arithmetic unit 40.

[0034] The communication unit 38 may be comprised of a communication interface interposed between the arithmetic unit 40 and the communication line 18. The communication unit 38 may be, for example, a NIC (Network Interface Card).

[0035] The arithmetic unit 40 may be composed of a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit). In other words, the arithmetic unit 40 may be composed of processing circuits.

[0036] The calculation unit 40 includes an information acquisition unit 46, a storage control unit 48, a display control unit 50, a regression curve acquisition unit 52, a first parameter acquisition unit 54, a second parameter acquisition unit 56, a success / failure determination unit 58, a third parameter acquisition unit 60, an information generation unit 62, and a parameter element determination unit 64. Each of the information acquisition unit 46, storage control unit 48, display control unit 50, regression curve acquisition unit 52, first parameter acquisition unit 54, second parameter acquisition unit 56, success / failure determination unit 58, third parameter acquisition unit 60, information generation unit 62, and parameter element determination unit 64 can be realized by the calculation unit 40 executing a program stored in the storage unit 42.

[0037] Furthermore, at least a portion of the information acquisition unit 46, the memory control unit 48, the display control unit 50, the regression curve acquisition unit 52, the first parameter acquisition unit 54, the second parameter acquisition unit 56, the success / failure determination unit 58, the third parameter acquisition unit 60, the information generation unit 62, and the parameter element determination unit 64 may be implemented by integrated circuits such as ASICs (Application Specific Integrated Circuits) and FPGAs (Field-Programmable Gate Arrays). Also, at least a portion of the information acquisition unit 46, the memory control unit 48, the display control unit 50, the regression curve acquisition unit 52, the first parameter acquisition unit 54, the second parameter acquisition unit 56, the success / failure determination unit 58, the third parameter acquisition unit 60, the information generation unit 62, and the parameter element determination unit 64 may be configured by electronic circuits including discrete devices.

[0038] The information acquisition unit 46 acquires various types of information from outside the calculation unit 40. The memory control unit 48 stores the various types of information in the memory unit 42. The display control unit 50 displays the various types of information on the display screen of the display unit 44. The regression curve acquisition unit 52 acquires a regression curve (e.g., a logistic regression curve) that shows the relationship between a physical quantity correlated with the number of cells and the culture time, based on the results of cell culture. The physical quantity correlated with the number of cells here may be glucose concentration, lactate concentration, etc., included in the sampling information, or it may be information acquired based on the sampling information. The first parameter acquisition unit 54 acquires a first parameter that includes multiple parameter elements included in the regression equation (e.g., a logistic function) that shows the regression curve. The second parameter acquisition unit 56 acquires a second parameter that includes multiple parameter elements based on the tangent line at the inflection point 72 (Figure 4) of the regression curve. The success / failure determination unit 58 inputs the parameter elements included in the first parameter, the parameter elements included in the second parameter, and the culture information into a machine learning model to determine the success or failure of the cell culture. The third parameter acquisition unit 60 acquires a third parameter containing multiple parameter elements by performing principal component analysis on the first and second parameters. The information generation unit 62 generates correlation information, which is information showing the relationship between one parameter element and other parameter elements for each of the multiple cell cultures performed. The parameter element determination unit 64 determines the parameter elements that contribute to the success or failure of the cell culture from among the multiple parameter elements included in the first parameter and the multiple parameter elements included in the second parameter.

[0039] The storage unit 42 may be constituted by a volatile memory (not shown) and a non-volatile memory (not shown). Examples of the volatile memory may include RAM (Random Access Memory) and the like. This volatile memory is used as a working memory for a processor, and temporarily stores data and the like necessary for processing or calculation. Examples of the non-volatile memory may include ROM (Read Only Memory), flash memory and the like. This non-volatile memory is used as a storage memory, and stores programs, tables, maps and the like. At least a part of the storage unit 42 may be provided in the processor, integrated circuit or the like as described above.

[0040] The display unit 44 may include, for example, a liquid crystal display, an organic EL display, or the like. The display unit 44 displays various display objects on a display screen in accordance with control by a display control unit 50.

[0041] [1-3 Server Device 16] The server device 16 may be constituted by a physical server or a cloud server. The server device 16 includes a processor and a memory (both not shown). The server device 16 acquires information from the cell culture device 12 via the communication line 18. The information acquired by the server device 16 is stored in the memory of the server device 16. The server device 16 transmits information to the information processing device 14 via the communication line 18.

[0042] [2 Cell Culture Information] N sets of cell culture information from previously performed culture rounds are stored in a memory of the control device 24 provided in the cell culture device 12. The cell culture information is acquired each time one round of cell culture is performed. The control device 24 transmits one set or multiple sets of cell culture information to at least one of the information processing device 14 and the server device 16 via the communication line 18. The information processing device 14 acquires cell culture information for each round from the control device 24 or the server device 16 via the communication line 18. The information processing device 14 can also acquire cell culture information through an operation performed by a user on the input unit 36. In the information processing device 14, the cell culture information is stored in the storage unit 42.

[0043] The cell culture information includes information on the used cells, the used liquid agents, and the like. This information includes, for example, information indicating each of cell species, cell type (adherent cells or suspension cells), cell source, cell origin, passage number, medium type, coating type, coating amount, coating grade, dissociation agent type, dissociation time, and buffer type.

[0044] The cell culture information includes information on the operation of the cell culture device 12. This information includes, for example, information indicating each of the medium discharge direction, the seeding method, and the flow velocity (or flow rate) of the pump 28.

[0045] The cell culture information includes information on cell culture conditions and cell culture results. This information includes, for example, information indicating each of the number of seeded cells, cell viability at seeding, the number of viable cells at seeding, seeding density, medium volume at seeding (mL), culture period (days), the number of harvested cells, cell viability at harvesting, the number of viable cells at harvesting, population doubling, doubling time, and number of population doublings. Each of population doubling, doubling time, and number of population doublings is calculated by the following formulas. Population doubling = Number of harvested cells / Number of seeded cells Doubling time = Log(2) × Culture period (hr) Number of population doublings = Log(Population doubling) / Log(2)

[0046] The cell culture information includes information on cell metabolism. This information includes, for example, information indicating each of gas module concentration, chemical module concentration, osmotic pressure, production rate of each chemical module (mmol / day), consumption rate of each chemical module (mmol / day), production amount of each module per cell (mmol / day / cell), and consumption amount of each module per cell (mmol / day / cell). The information on cell metabolism is acquired based on sampling information. The gas module measures pH, carbon dioxide, and oxygen. The chemical module measures glucose, lactic acid, glutamine, glutamic acid, ammonium ions, sodium ions, potassium ions, and calcium ions.

[0047] Cell culture information includes information indicating the success or failure of cell culture (culture success / failure information). This information includes information indicating the user's judgment of success or failure (also called primary judgment information). The user's judgment of culture success or failure is based on information related to the quality of the obtained cells, such as the number of cells recovered and the viability of the cells at the time of recovery. In other words, the success or failure of the culture is determined based on the user's own subjective judgment criteria. The judgment may vary depending on the user, and it may be difficult to make a stable judgment of culture success or failure. As will be described later, this information may be modified by the information processing device 14.

[0048] [3 Machine Learning Model] The memory of the control device 24 provided in the cell culture apparatus 12 stores a machine learning model used by the success / failure determination unit 58 of the calculation unit 40. The machine learning model is pre-trained to output the result of determining the success or failure of cell culture. As the machine learning model, decision trees that can confirm classification conditions, random forests, LightBGM, XGBooster, and other decision tree methods can be used. In addition to decision tree methods, neural networks, logistic regression, and simple Bayesian classification can also be used. In this specification, an embodiment using the decision tree model 68 shown in Figure 2 as the machine learning model will be described.

[0049] Figure 2 shows an example of a decision tree model 68. Based on the input information (a portion of the information contained in a single cell culture), the decision tree model 68 outputs either a result indicating whether the cell culture was successful or unsuccessful, or a result indicating that success or failure is not possible.

[0050] The decision tree model 68 shown in Figure 2 has three criteria: "doubling time < 40h", "LagPhase < 50h", and "parameter c < 3.0". Note that the criteria that the decision tree model 68 may have are not limited to these. The depth of the decision tree model 68 shown in Figure 2 is 2. Note that the depth of the decision tree model 68 is not limited to this. Figure 2 shows a simplified decision tree model 68.

[0051] It is preferable to use information related to cell doubling (doubling information) as input to the decision tree model 68. This information includes, in addition to the doubling time shown in Figure 2, the number of doublings, the number of doubling cycles, etc. Including cell doubling information in the decision tree model 68 simplifies (makes easier) the classification of decision boundaries in the decision tree.

[0052] The decision tree model 68, as shown in Figure 2, comprises at least two layers of decision criteria. Furthermore, the decision tree model 68 includes information regarding the doubling of at least one cell. Additionally, the decision tree model 68 comprises decision criteria including at least one first parameter and at least one second parameter.

[0053] [4 Functions of the Information Processing Device 14] [4-1 Analysis Processing] Figure 3 is a flowchart of the analysis processing performed by the Information Processing Device 14. As described above, the Information Processing Device 14 can perform cell culture information analysis on behalf of the user. The analysis processing performed by the Information Processing Device 14 will be explained below using Figure 3. The storage unit 42 provided in the Information Processing Device 14 stores cell culture information for N past cycles. The cell culture information for N past cycles is the population data used for analysis. The population data used for analysis may be data that has been extracted in advance by any or a combination of cell culture information. For example, the population data may be data that has been classified in advance by information such as cell type and cell type (adherent cells or suspension cells) as cell culture information.

[0054] In step S1, the information acquisition unit 46 adds 1 to the value (c) of the counter. The initial value of the counter is zero (c = 0).

[0055] In step S2, the information acquisition unit 46 acquires cell culture information for the c-th cell culture from among the N cell culture information. "c" is the value of the counter.

[0056] In step S3, the regression curve acquisition unit 52 acquires a regression curve showing the relationship between a physical quantity correlated with the number of cells and the culture time, based on the results of cell culture. For example, the cell culture information acquired by the information acquisition unit 46 includes time-series metabolic information (production rate of each chemical module, consumption rate of each chemical module, etc.). Metabolic information corresponds to a physical quantity correlated with the number of cells. Below, an embodiment will be described using the production rate of lactate from the metabolic information. The regression curve acquisition unit 52 acquires a logistic regression curve based on the time-series production rate of lactate.

[0057] Figure 4 is a graph showing the relationship between culture time and lactic acid production rate. The lactic acid production rate in Figure 4 is on a logarithmic scale. Curve 70 in Figure 4 is a logistic regression curve showing the relationship between lactic acid production rate (a physical quantity correlated with the number of cells) and culture time. The regression curve acquisition unit 52 acquires curve 70 by regression of the time-series lactic acid production rate information contained in the cell culture information using a logistic function.

[0058] In step S4, the first parameter acquisition unit 54 acquires a first parameter that includes multiple parameter elements. The logistic regression curve is represented by the logistic function in equation (1) below. In equation (1) below, logN represents the logarithm of the lactic acid production rate, and t represents the culture time.

[0059] The logistic function in equation (1) above includes four parameter elements: parameter a, parameter b, parameter c, and parameter d. These four parameter elements are called the first parameters. The first parameter acquisition unit 54 acquires parameters a, b, c, and d included in the logistic function. The storage control unit 48 stores the values ​​of each of the four parameter elements acquired by the first parameter acquisition unit 54 in the storage unit 42.

[0060] In step S5, the second parameter acquisition unit 56 acquires a second parameter that includes multiple parameter elements. In Figure 4, the straight line 74 represents the tangent line at the inflection point 72 of the curve 70. The second parameter acquisition unit 56 acquires the straight line 74 and acquires three parameter elements, capA, argmax_dt, and LagPhase, based on the straight line 74. These three parameter elements are called the second parameter.

[0061] capA indicates the maximum metabolic rate. capA corresponds to the vertical axis coordinate value (lactic acid production rate) at the inflection point 72 of curve 70. argmax_dt indicates the culture time at which the maximum metabolic rate is reached. argmax_dt corresponds to the horizontal axis coordinate value (culture time) at the inflection point 72 of curve 70. LagPhase indicates the culture time at which the cell transitions from the LAG phase (induction phase) to the LOG phase (logarithmic growth phase). LagPhase corresponds to the horizontal axis coordinate value (culture time) at the intersection point 78 of the baseline 76 of curve 70 and the straight line 74. The value of baseline 76 corresponds to the value of parameter d among the first parameters. The memory control unit 48 stores the values ​​of each of the three parameter elements acquired by the second parameter acquisition unit 56 in the memory unit 42.

[0062] In step S6, the success / failure determination unit 58 determines the success or failure of the c-th cell culture. Here, the success / failure determination unit 58 inputs at least one piece of information from the parameter elements included in the first parameter, the parameter elements included in the second parameter, and the culture information into the decision tree model 68 to determine the success or failure of the cell culture. When the decision tree model 68 shown in Figure 2 is used, the success / failure determination unit 58 inputs parameter c included in the first parameter, LagPhase included in the second parameter, and the doubling time included in the cell culture information into the decision tree model 68. The success / failure determination unit 58 obtains the output of the decision tree model 68 (either success, failure, or inability to determine success or failure) as the result of determining the success or failure of the cell culture.

[0063] In step S7, the memory control unit 48 stores information indicating the determination result by the success / failure determination unit 58 in the memory unit 42. The memory control unit 48 associates the information indicating the determination result with the cell culture information and stores it in the memory unit 42. The cell culture information may include information indicating the user's determination result regarding the success or failure of cell culture (primary determination information). In such cases, the memory control unit 48 may store information indicating the determination result by the success / failure determination unit 58 (also called secondary determination information) in the memory unit 42, separately from the information indicating the user's determination result. Alternatively, the memory control unit 48 may update the information indicating the user's determination result with the information indicating the determination result by the success / failure determination unit 58.

[0064] In step S8, the information acquisition unit 46 determines whether the counter value (c) is N or greater. That is, it determines whether the acquisition of the first and second parameters has been completed for each of the N cell cultures. If the acquisition of the first and second parameters has been completed for each of the N cell cultures, i.e., if the counter value (c) is N or greater (step S8: YES), the process proceeds to step S9. On the other hand, if the acquisition of the first and second parameters has not been completed for some of the cell cultures, i.e., if the counter value (c) is less than N (step S8: NO), the process returns to step S1.

[0065] In step S9, the third parameter acquisition unit 60 acquires a third parameter including multiple parameter elements as follows. When step S8 is completed, each cell culture information stored in the storage unit 42 is associated with seven-dimensional parameter elements (parameters a to d, capA, argmax_dt, LagPhase). Here, the third parameter acquisition unit 60 performs a process to reduce the seven-dimensional parameter elements to six-dimensional or less. For example, the third parameter acquisition unit 60 generates four principal components (also called PCAs) by performing principal component analysis based on the seven-dimensional parameter elements in the cell culture information for N times. The four principal components consist of a first principal component (also called PCA0), a second principal component (also called PCA1), a third principal component (also called PCA2), and a fourth principal component (also called PCA3).

[0066] Figures 5A to 5D are graphs showing the contribution (coefficient) of the first and second parameters to each principal component. Figure 5A shows how much each parameter element of the first and second parameters is reflected in the first principal component (PCA0). Figure 5B shows how much each parameter element of the first and second parameters is reflected in the second principal component (PCA1). Figure 5C shows how much each parameter element of the first and second parameters is reflected in the third principal component (PCA2). Figure 5D shows how much each parameter element of the first and second parameters is reflected in the fourth principal component (PCA3). The ratios indicated at the top of each of Figures 5A to 5D indicate the contribution rate of the principal component. The third parameter acquisition unit 60 acquires the contribution (coefficient) of each parameter element shown in Figures 5A to 5D.

[0067] Furthermore, the third parameter acquisition unit 60 calculates the score for each principal component based on the values ​​of each parameter element of the first and second parameters and the contribution (coefficient) of each parameter element. For example, the third parameter acquisition unit 60 multiplies the value of each parameter element for one cell culture by the corresponding coefficient of the first principal component, and the sum of the multiplied values ​​is taken as the score for the first principal component for one cell culture. The third parameter acquisition unit 60 calculates the score for the second principal component, the score for the third principal component, and the score for the fourth principal component in the same manner as the score for the first principal component. The third parameter acquisition unit 60 calculates the score for each principal component for each cell culture.

[0068] As described above, the third parameter acquisition unit 60 calculates the score for each of the four principal components corresponding to the seven parameter elements for each cell culture cycle. In other words, the third parameter acquisition unit 60 reduces the seven-dimensional parameter elements to four-dimensional parameter elements (scores of the principal components). These four parameter elements are called the third parameters.

[0069] In step S9, the memory control unit 48 causes the memory unit 42 to store information indicating each of the four parameter elements acquired by the third parameter acquisition unit 60.

[0070] In step S10, the information generation unit 62 generates correlation information that associates one parameter element with another parameter element from among the multiple parameter elements included in each of the first parameter, second parameter, and third parameter. As an example, the information generation unit 62 generates correlation information that associates the first principal component and the second principal component included in the third parameter. As another example, the information generation unit 62 generates correlation information that associates parameter a included in the first parameter with capA included in the second parameter. The information generation unit 62 can also generate correlation information that associates two other parameter elements. The information generation unit 62 may also generate correlation information that associates three or more parameter elements. The storage control unit 48 stores the correlation information in the storage unit 42.

[0071] In step S11, the parameter element determination unit 64 determines the parameter elements that contribute to cell culture. For example, the parameter element determination unit 64 determines one or more parameter elements that contribute to cell culture based on the correlation information generated by the information generation unit 62. An example of the processing performed by the parameter element determination unit 64 is described below.

[0072] Figure 6 is a scatter plot showing the relationship between the first principal component (PCA0) and the second principal component (PCA1). Figure 7 is a scatter plot showing the relationship between the first principal component (PCA0) and the third principal component (PCA2). Figure 8 is a scatter plot showing the relationship between the first principal component (PCA0) and the fourth principal component (PCA3). Figure 9 is a scatter plot showing the relationship between the second principal component (PCA1) and the third principal component (PCA2). Figure 10 is a scatter plot showing the relationship between the second principal component (PCA1) and the fourth principal component (PCA3). Figure 11 is a scatter plot showing the relationship between the third principal component (PCA2) and the fourth principal component (PCA3). The scatter plots shown in Figures 6 to 11 correspond to correlation information between two principal components (parameter elements). Each point plotted in each scatter plot shows the relationship between two principal components for a single cell culture. Each point plotted in each scatter plot is associated with information regarding the success or failure of the cell culture. In the scatter plots shown in Figures 6 to 11, points associated with a success result are shown as circles, and points associated with a failure result are shown as triangles. Note that points associated with an inability to determine success or failure are not shown in the scatter plots shown in Figures 6 to 11.

[0073] The parameter element determination unit 64 selects a scatter plot from among those shown in Figures 6 to 11 that can create a boundary between the cluster of points associated with the success determination result and the cluster of points associated with the failure determination result. For example, if the parameter element determination unit 64 selects the scatter plot shown in Figure 11 (relationship between the score of the third principal component and the score of the fourth principal component), it performs the following processing.

[0074] The parameter element determination unit 64 performs analysis on the third principal component and the fourth principal component. The parameter element determination unit 64 identifies the parameter elements of the first and second parameters for the third principal component whose contribution (coefficient) is a positive value. In Figure 5C, the parameter elements with a positive contribution are parameter b, parameter c, capA, and LagPhase. Here, these parameter elements are referred to as the first determination parameters. Similarly, the parameter element determination unit 64 identifies the parameter elements of the first and second parameters for the fourth principal component whose contribution (coefficient) is a positive value. In Figure 5D, the parameter elements with a positive contribution are parameter b, argmax_dt, and LagPhase. Here, these parameter elements are referred to as the second determination parameters. The parameter element determination unit 64 identifies the parameter elements that are included in both the first determination parameters and the second determination parameters. In the above example, the parameter elements included in both the first and second determination parameters are parameter b and LagPhase. Through the above process, the parameter element determination unit 64 determines that parameter b and LagPhase are parameter elements that contribute to cell culture.

[0075] The parameter element determination unit 64 may determine the parameter elements that contribute to cell culture based on a single parameter element, or it may determine them based on multiple parameter elements.

[0076] In step S12, the display control unit 50 displays the determination result from the parameter element determination unit 64 on the display screen of the display unit 44. In the above example, the display control unit 50 displays on the display screen of the display unit 44 the determination result that parameter b and LagPhase are parameter elements that contribute to cell culture.

[0077] [4-2 Support Processing] Figure 12 is a flowchart of the support processing performed by the information processing device 14. As described above, the information processing device 14 can support users in analyzing cell culture information. The support processing performed by the information processing device 14 will be explained below using Figure 12. Note that the processing in steps S21 to S30 shown in Figure 12 is the same as the processing in steps S1 to S10 shown in Figure 3. For this reason, the explanation of the processing in steps S21 to S30 will be omitted below.

[0078] In step S31, the display control unit 50 displays the correlation information on the display screen of the display unit 44. For example, the display control unit 50 displays each of the scatter plots shown in Figures 6 to 11 on the display screen of the display unit 44. The display control unit 50 may display each scatter plot individually, or it may display multiple scatter plots in a list. Furthermore, the display control unit 50 displays the graphs of the contribution (coefficients) shown in Figures 5A to 5D on the display screen of the display unit 44. This allows the user to visually confirm the correlation information of the two principal components. The user can then perform the processing performed by the parameter element determination unit 64 (step S11 in Figure 3).

[0079] Figure 13 is a scatter plot showing the relationship between parameter a and capA. The scatter plot shown in Figure 13 corresponds to correlation information between parameter a (parameter element) and capA (parameter element). The display control unit 50 may display the scatter plot shown in Figure 13 on the display screen of the display unit 44. The display control unit 50 may also display a scatter plot showing the relationship between two of the multiple parameter elements on the display screen of the display unit 44. This allows the user to perform further analysis.

[0080] According to the embodiments described above, it is possible to perform analyses of cell culture from a new perspective. According to the embodiments described above, it is not necessary to conduct experiments to determine the optimal cell culture conditions.

[0081] Furthermore, according to the embodiments described above, information for determining optimal cell culture conditions can be provided to inexperienced users. In other words, according to the embodiments described above, cell culture information can be analyzed regardless of the user's experience. That is, according to the embodiments described above, it is possible to suppress variations in cell quality identified by the analysis of cell culture information and variations in costs associated with the analysis of cell culture information.

[0082] Furthermore, according to the above-described embodiment, the success or failure of previously performed cell culture can be determined regardless of the user's experience. Moreover, according to the above-described embodiment, if there is an error in the user's judgment regarding the success or failure of cell culture, the error can be corrected.

[0083] [5. Examples of Cell Culture] Specific examples of cell culture are described below.

[0084] [5-1 Example of Adherent Cell Culture] The inventors cultured adherent cells. The adherent cells used were human adipose-derived stem cells (LONZA, PT-5006). The cell culture apparatus 12 used was an automated culture apparatus (Quantum® (Terumo BCT)) equipped with a bioreactor 26.

[0085] The cell culture apparatus 12 primed the circuit with PBS (phosphate-buffered saline) (1×) and coated the bioreactor 26 overnight with a coating solution containing fibronectin at a final concentration of 5 mg / 100 mL. Subsequently, the cell culture apparatus 12 replaced the solution in the circuit from PBS (1×) to MSCGM2 medium. The cell culture apparatus 12 supplied gas (a mixed gas of 75% nitrogen, 20% oxygen, and 5% carbon dioxide) and conditioned the medium for 1 to 4 hours. Using a cell suspension prepared in advance with MSCGM2, the cell culture apparatus 12 seeded the cell suspension on the IC side of the bioreactor 26 after gas conditioning was completed, and the cells (human adipose mesenchymal stem cells) adhered to the bioreactor 26 overnight. Subsequently, the cell culture apparatus 12 supplied MSCGM2 medium according to the flow rate shown in Figure 14 and performed cell culture. After the culturing was complete, the cell culture apparatus 12 detached the cells from the bioreactor 26 using trypsin solution and collected the detached cells.

[0086] The inventors collected 1 mL of culture medium from the sampling port of the sampling unit 30 and measured the glucose concentration and lactate concentration using the i-STAT® 1 Analyzer. Figure 15 is a graph showing the relationship between culture time and glucose concentration and the relationship between culture time and lactate concentration. Figure 16 is a graph showing the relationship between culture time and the metabolic rate of cells (glucose consumption rate and lactate production rate). Figure 17 is a table showing the number of seeded cells, the number of harvested cells, the doubling time, the number of doublings, and the number of doubling cycles.

[0087] The inventors calculated glucose consumption based on the following formula (2) and lactic acid production based on the following formula (3). Glc: Glucose concentration (mol / ml) Q: Inlet rate (ml / min) Glc ｓｕｐ : Concentration of glucose supplied to cell culture circuit 22 (mol / ml) Lac: Lactate concentration (mol / ml) Lac ｓｕｐ : Concentration of lactate supplied to cell culture circuit 22 (molecule / ml) M Ｌａｃ : Lactic acid production rate (mmol / cell) X: Number of cells MＧｌｃ : Glucose consumption rate (moles / cell) Δt: Calculation interval (min) V: Priming volume of bioreactor 26 (ml)

[0088] [5-2 Example of culture of suspension cells] The inventors cultured suspension cells. The suspension cells used were Jurkat cells (human cell leukemia-derived cells) (ATCC E6-1). The cell culture apparatus 12 used was an automated culture apparatus (Quantum) equipped with a bioreactor 26. Unless otherwise specified, the procedure for culturing suspension cells is the same as the procedure for culturing adherent cells.

[0089] The cell culture apparatus 12 supplied complete medium to the IC side of the bioreactor 26 and basal medium to the EC side of the bioreactor 26. The complete medium was RPMI1640 medium containing 5% FBS (Fetal Bovine Serum). The basal medium was RPMI1640 medium. In the culture of suspension cells, the bioreactor 26 was not coated with a coating solution. The cell culture apparatus 12 supplied MSCGM2 medium according to the flow rate shown in Figure 18 and performed cell culture.

[0090] The inventors collected 0.2 mL of culture medium from the sampling port of the sampling unit 30 and measured the glucose concentration and lactate concentration using an i-STAT1 analyzer. Figure 19 is a graph showing the relationship between culture time and glucose concentration and the relationship between culture time and lactate concentration. Figure 20 is a graph showing the relationship between culture time and the metabolic rate of cells (glucose consumption rate and lactate production rate). Figure 21 is a table showing the number of seeded cells, the number of harvested cells, the doubling time, the number of doublings, and the number of doubling cycles.

[0091] The inventors calculated glucose consumption based on equation (2) above and lactic acid production based on equation (3) above.

[0092] While this disclosure has been described in detail, it is not limited to the individual embodiments described above. These embodiments can be added, replaced, modified, partially deleted, etc., in any way that does not depart from the gist of this disclosure or from the intent of this disclosure derived from the claims and their equivalents. These embodiments can also be implemented in combination. For example, the order of operations and processes in the embodiments described above are given as examples only and are not limited thereto. The same applies when numerical values ​​or mathematical formulas are used in the description of the embodiments described above.

Claims

1. An information processing device comprising: a regression curve acquisition unit that acquires a regression curve showing the relationship between a physical quantity correlated with the number of cells and the culture time based on the results of cell culture; a first parameter acquisition unit that acquires a first parameter including a plurality of parameter elements included in the regression equation showing the regression curve; a second parameter acquisition unit that acquires a second parameter including a plurality of parameter elements based on the tangent line at the inflection point of the regression curve; wherein the first parameter acquisition unit acquires the first parameter for each of the multiple cell cultures performed; the second parameter acquisition unit acquires the second parameter for each of the multiple cell cultures performed; a third parameter acquisition unit that acquires a third parameter including a plurality of parameter elements by performing principal component analysis on the first parameter and the second parameter; and an information generation unit that generates correlation information which is information showing the relationship between one parameter element and other parameter elements for each of the multiple cell cultures performed.

2. An information processing device according to claim 1, wherein each parameter element included in the third parameter is a score of each principal component generated by the principal component analysis.

3. An information processing apparatus according to claim 1, comprising a display control unit for displaying the correlation information on a display screen.

4. An information processing device according to claim 3, wherein the display control unit causes the scatter plot, which is the correlation information, to be displayed on the display screen.

5. An information processing device according to claim 4, wherein the display control unit displays on the display screen, in addition to the scatter plot, the contribution of each parameter element of the first parameter and the contribution of each parameter element of the second parameter in each principal component generated by the principal component analysis.

6. An information processing device according to claim 1, comprising a parameter element determination unit that determines a parameter element that contributes to the success or failure of cell culture from among a plurality of parameter elements included in the first parameter and a plurality of parameter elements included in the second parameter, wherein the third parameter for each of the multiple cell cultures performed and the culture success or failure information, which is information indicating the success or failure of cell culture, are associated with each other, and the parameter element determination unit determines the parameter element that contributes to the success or failure of cell culture based on the contribution of each parameter element of the first parameter and each parameter element of the second parameter in each principal component generated by the principal component analysis, the scatter plot which is correlation information, and the culture success or failure information.

7. An information processing method comprising: a regression curve acquisition step of obtaining a regression curve showing the relationship between a physical quantity correlated with the number of cells and the culture time based on the results of cell culture; a first parameter acquisition step of obtaining a first parameter including a plurality of parameter elements included in the regression equation showing the regression curve; a second parameter acquisition step of obtaining a second parameter including a plurality of parameter elements based on the tangent line at the inflection point of the regression curve, wherein in the first parameter acquisition step, the first parameter is obtained for each of the multiple cell cultures performed; in the second parameter acquisition step, the second parameter is obtained for each of the multiple cell cultures performed; a third parameter acquisition step of obtaining a third parameter including a plurality of parameter elements by performing principal component analysis on the first parameter and the second parameter; and an information generation step of generating correlation information which is information showing the relationship between one parameter element and other parameter elements for each of the multiple cell cultures performed.

8. An information processing method according to claim 7, wherein each parameter element included in the third parameter is the score of each principal component generated by the principal component analysis.

9. An information processing method according to claim 7, comprising a display step of displaying the correlation information on a display screen.

10. An information processing method according to claim 9, wherein the display step involves displaying the scatter plot, which is the correlation information, on the display screen.

11. An information processing method according to claim 10, wherein in the display step, in addition to the scatter plot, the contribution of each parameter element of the first parameter and the contribution of each parameter element of the second parameter in each principal component generated by the principal component analysis are displayed on the display screen.

12. An information processing method according to claim 7, comprising a determination step of determining a parameter element that contributes to the success or failure of cell culture from among a plurality of parameter elements included in the first parameter and a plurality of parameter elements included in the second parameter, wherein the third parameter for each of the multiple cell cultures performed and the culture success or failure information, which is information indicating the success or failure of cell culture, are associated with each other, and the determination step determines the parameter element that contributes to the success or failure of cell culture based on the contribution of each parameter element of the first parameter and each parameter element of the second parameter in each principal component generated by the principal component analysis, the scatter plot which is correlation information, and the culture success or failure information.

13. A program for causing a computer to execute the information processing method described in any one of claims 7 to 12.

Images