Configuration method, configuration device, configuration program, and recording medium

Abstract

One form of the present invention provides a construction method, a construction device, and a construction program with which it is possible, with size constraints for a limited set of culture media, to construct a culture media panel having high compatibility with cells having various characteristics. In one form of the present invention, a construction method for a culture media panel comprising a plurality of culture media optimizes cell culture results, and in said method, a processor executes: a region setting step for setting regions in a space for culture medium compositions; a characteristic enumeration step for enumerating a plurality of cell characteristics; a prediction step for predicting, on the basis of the cell characteristics of a target cell, culture results for the target cell in a provided culture medium composition; a search and enumeration step for searching for and enumerating culture medium compositions for which the cell culture results are very high values for each cell characteristic included in the plurality of cell characteristics; and a construction step for constructing a culture media panel using the culture medium composition candidates associated with the very high values.

Description

Configuration method, configuration device, and configuration program

[0001] The present invention relates to a method, an apparatus, and a program for constructing a culture medium panel consisting of a plurality of culture media.

[0002] Recently, with the spread of gene editing technology, the diversification of various cells has progressed, such as CHO (Chinese Hamster Ovary) cells that produce antibody drugs, HEK (Human Embryonic Kidney) cells that produce viruses for gene therapy, and iPS (Induced Pluripotent Stem) cells for regenerative medicine. Generally, there are many cases where cell culture conditions are optimized to increase cell proliferation and the production of various products, but there are also many cases where it is desired to optimize the culture medium for specific cells.

[0003] For example, Patent Document 1 describes the construction of a functional enviromics map that indicates the strength of activation or inhibition of basic cellular functions, and the development of an optimized cell culture medium formulation (culture medium composition) using the functional enviromics map. Furthermore, Patent Document 2, for example, describes the production of a (new) culture medium based on the culture object (cells), the culture evaluation index, and the predicted parameters of the culture medium production conditions.

[0004] Patent Publication No. 2014-503220 International Publication No. 2021 / 049044

[0005] However, developing a culture medium specifically for a particular cell type (hereafter referred to as a dedicated culture medium) requires time and costs. In contrast, the method of selecting an appropriate culture medium through a small number of repeated experiments is limited by the simplified development of a dedicated culture medium, and there is also the problem that simply using any suitable medium for the time being leaves open the question of how reliable the results can be.

[0006] Therefore, one possible method is to prepare a "set" of multiple suitable culture media (hereafter referred to as a "culture media panel") in advance, conduct a culture experiment once (or a few times) using the culture media panel, and if any of the culture experiments produce good results, select that medium. In this way, even when developing a dedicated culture medium, starting from a medium included in the culture media panel can shorten the development time and cost.

[0007] Therefore, "constructing a culture medium panel that can accommodate a variety of cells" is an important issue.

[0008] A common solution to this problem is to construct a panel by appropriately adding proven culture media based on experience and knowledge from past culture experiments. However, such a common solution faces difficult constraints: (1) because cell information cannot be obtained in advance, it is impossible to predict which cells should be used, and as mentioned above, the diversity of cell types is increasing; and (2) because confirmation requires experiments, the size of the culture media panel (the number of culture media included in a culture media set) cannot be made very large. Therefore, as the diversity of cell types continues to increase, new technological developments are required, including alternative methods for when a culture medium that was previously effective no longer works. However, conventional technologies such as those described in Patent Documents 1 and 2 have had difficulty responding to such situations.

[0009] The present invention has been made in consideration of the above circumstances, and aims to provide a construction method, construction device, and construction program that can construct a culture medium panel that is small in size and highly compatible with cells with diverse characteristics.

[0010] In order to achieve the above-mentioned object, a configuration method according to a first aspect of the present invention is a method for configuring a culture medium panel consisting of a plurality of culture media to optimize cell culture results, the method comprising the steps of: a region setting step for setting a region in a culture medium composition space; a characteristic enumeration step for enumerating a plurality of cellular characteristics; a prediction step for predicting the culture result of a target cell in a given culture medium composition based on the cellular characteristics of the target cell; a search and enumeration step for searching for and enumerating, for each cellular characteristic included in the plurality of cellular characteristics, medium compositions at which the cell culture result will be a local maximum; and a configuration step for configuring a culture medium panel from candidate medium compositions corresponding to the local maximum; in the search and enumeration step, the processor extracts candidate medium compositions from the region in the culture medium composition space, and predicts the culture result of a cell having a cellular characteristic included in the plurality of cellular characteristics in a prediction step, and searches for and determines candidate medium compositions that will give a local maximum from the predicted culture result.

[0011] According to the first aspect, the culture medium composition that maximizes the cell culture result is searched for and listed for each cell characteristic included in the plurality of cell characteristics, and a culture medium panel is constructed from candidate culture medium compositions that correspond to the maximum values ​​of the culture result. This makes it possible to construct a culture medium panel that is small in size (the number of culture media included in the culture medium panel is small) and highly compatible with cells of diverse characteristics.

[0012] In the first aspect and each of the following aspects, the term "target cell" refers to a "cell that has been given specific cellular characteristics." Note that the characteristics of a "target cell" must be clearly defined, but the "target cell" does not necessarily mean a "known" cell in the sense of being widely available or having been created; it also includes "new" cells that are about to be created, i.e., cells that have not yet been created, and "virtual cells with virtual cellular characteristics." However, "unknown cells" in the sense of having unclear characteristics are not included in the target cell.

[0013] Furthermore, in the first aspect and each of the following aspects, "a medium of a given composition" means a "medium of a specific composition" or a "medium of any composition," and the amount of each component in the composition is clearly defined. This "medium of a given composition" is not limited to a "known" medium in the sense of being commonly used or already being produced, but may also be a "novel" medium that is about to be produced, i.e., has not yet been produced. However, an "unknown" medium in which the amount of components is unclear is not included in the "medium of a given composition."

[0014] In the first embodiment and each of the following embodiments, the culture result being a "maximum value" means that "considering the characteristics of each individual cell, it is preferable to be as close to the maximum value as possible, but considering compatibility with multiple cells, it does not have to be the exact maximum value and may be close to the maximum value."

[0015] In the configuration method according to the second aspect, in the first aspect, the processor configures a culture medium panel that broadens the distribution of candidate culture medium compositions in the culture medium composition space in the configuration step. By selecting media that configure the culture medium panel from a variety of media, it is possible to accommodate unknown cells with characteristics that deviate somewhat from the expected cell characteristics, and the second aspect specifies specific aspects of this "various media."

[0016] In the configuration method according to the third aspect, in the second aspect, the processor configures, in the configuration step, a culture medium panel in which the correlation between candidate culture medium compositions in the culture medium composition space is reduced, a culture medium panel in which the distance between candidate culture medium compositions in the culture medium composition space is increased, or a culture medium panel in which the diversity defined by the information gain of candidate culture medium compositions in the culture medium composition space is increased. The third aspect specifies one aspect of the configuration method for the "culture medium panel in which the distribution of candidate culture medium compositions in the culture medium composition space is broadened" in the second aspect.

[0017] A fourth aspect of the present invention is a configuration method according to any one of the first to third aspects, wherein the processor weights the candidate medium compositions based on the frequency of appearance of the cell characteristics in the configuration step, and configures the culture medium panel based on the weighting results. According to the fourth aspect, an appropriate culture medium panel can be configured based on the frequency of appearance of the cell characteristics.

[0018] A fifth aspect of the configuration method is any one of the first to fourth aspects, in which the processor executes an evaluation step of evaluating the property difference between cells, and a generation step of generating a virtual cell property based on the evaluation while controlling the virtual cell property, such that the difference from the cell property of a known cell group satisfies a constraint. When enumerating multiple cell properties in the property enumeration step, one or more of the generated virtual cell properties are included. As specified in the fifth aspect, in the present invention, when enumerating multiple cell properties, virtual cell properties may be generated and added to the known cell properties. Note that the "known cell(s)" in the fifth aspect refers to cells that actually exist and are available. Since there is no point in considering impossible cell properties, the purpose is to create virtual cell properties by taking into account the cell properties of the "known cell(s)" that actually exist.

[0019] A sixth aspect of the present invention is the configuration method of the fifth aspect, wherein the processor sets the constraints based on characteristic differences between cells in a known cell group in the generation step. The sixth aspect defines one aspect of setting the constraints.

[0020] A seventh aspect of the invention relates to a configuration method for any one of the first to sixth aspects, in which the processor divides the culture medium panel into subsets and assigns at least one known cell lineage to one of the subsets. Attempting to configure a culture medium panel that corresponds to all anticipated cell lineages may result in a large size of the culture medium panel or in a coarse selection of culture medium compositions (i.e., medium compositions that produce a maximum or near-maximum culture result may not be sufficiently covered (i.e., may not be included in the culture medium panel)). In such cases, configuring the culture medium panel as in the seventh aspect makes it possible to configure a more closely matched panel (a culture medium panel composed of candidate culture medium compositions that sufficiently cover medium compositions that produce a maximum or near-maximum culture result) while keeping the size of the culture medium panel small.

[0021] The construction method according to the eighth aspect is the same as that of the seventh aspect, in which the processor divides the culture medium panel into subsets, and includes a subset that is not dependent on a specific cell lineage. The eighth aspect defines one aspect of constructing the subsets.

[0022] A construction method according to a ninth aspect is any one of the first to eighth aspects, in which the processor constructs a culture medium panel using a submodular optimization method in the construction step. The ninth aspect defines one aspect of a method for constructing a culture medium panel. "Submodular optimization" is an optimization method that utilizes the submodularity of a set function, and "submodularity" is a concept corresponding to convexity or concavity (in a continuous function) defined for a set, and is known to be applicable to combinatorial optimization.

[0023] A tenth aspect of the present invention is a configuration method according to any one of the first to ninth aspects, wherein the processor expresses at least one of a prediction means in the prediction step, a prediction function for an unknown cell, and a difference between cell characteristics by a cell simulation including a metabolic pathway model. The tenth aspect specifically defines one method for predicting culture results.

[0024] A configuration method according to an eleventh aspect is any one of the first to tenth aspects, wherein in the prediction step, the processor predicts a biological behavior quantity of the target cell based on a given medium composition and culture conditions, modifies the cellular environment of the target cell based on the predicted biological behavior quantity, and repeats the prediction of the biological behavior quantity and the modification of the cellular environment based on the modified cellular environment, thereby predicting a culture result. The eleventh aspect specifically defines one method of predicting a culture result.

[0025] A configuration method according to a twelfth aspect is the eleventh aspect, in which the processor predicts, as biological behavior quantities, changes in the total number of cells, changes in the amount of substances and substances produced by the target cells over time, and changes in the amount of substances and substances contained in the culture environment of the target cells over time, and changes the total number of the target cells and the amount of substances produced by the target cells as the cellular environment. The twelfth aspect further specifies the prediction according to the eleventh aspect.

[0026] A thirteenth aspect of the present invention relates to any one of the first to twelfth aspects, wherein the target cells are CHO cells, HEK cells, or iPS cells. The thirteenth aspect specifically defines an example of the target cells.

[0027] A fourteenth aspect of the present invention is a method according to any one of the first to thirteenth aspects, wherein the culture results are one or more of the amount of antibody or virus produced by the target cells, the proliferation of the target cells, and the success rate of inducing differentiation of the target cells into a specific tissue. The fourteenth aspect defines an example of the culture results. In the present invention, it is preferable to use culture results according to the type of target cells.

[0028] In order to achieve the above-mentioned object, a configuration device according to a fifteenth aspect of the present invention is a configuration device for a culture medium panel consisting of a plurality of culture media that optimizes cell culture results, and includes a processor. The processor executes the following steps: a setting process for setting a region of a culture medium composition space; an enumeration process for enumerating a plurality of cellular characteristics; a prediction process for predicting the culture results of a target cell in a given culture medium composition based on the cellular characteristics of the target cell; a search and enumeration process for searching for and enumerating, for each cellular characteristic included in the plurality of cellular characteristics, medium compositions that will produce a maximum value in the cell culture result; and a configuration process for constructing a culture medium panel from candidate medium compositions that correspond to the maximum value. In the search and enumeration process, candidate medium compositions are extracted from the region of the culture medium composition space, and a prediction process is used to predict the culture results of a cell having a cellular characteristic included in the plurality of cellular characteristics, and a candidate medium composition that will produce a maximum value is searched for and determined from the predicted culture results.

[0029] According to the fifteenth aspect, it is possible to construct a culture medium panel that is small in size and highly compatible with cells with diverse characteristics, as in the first aspect. Note that the component device according to the fifteenth aspect may have the same configuration as the second to fourteenth aspects.

[0030] In order to achieve the above-mentioned object, a configuration program according to a sixteenth aspect of the present invention is a configuration program that causes a processor to execute a configuration method for a culture medium panel consisting of a plurality of culture media that optimizes cell culture results. The configuration method includes an area setting step of setting an area in a culture medium composition space, a characteristic enumeration step of enumerating a plurality of cellular characteristics, a prediction step of predicting the culture results of a target cell in a given culture medium composition based on the cellular characteristics of the target cell, a search and enumeration step of searching for and enumerating, for each cellular characteristic included in the plurality of cellular characteristics, medium compositions that will produce a maximum value in the cell culture result, and a configuration step of configuring a culture medium panel from candidate medium compositions that correspond to the maximum value. In the search and enumeration step, candidate medium compositions are extracted from the area in the culture medium composition space, and a prediction step predicts the culture results of a cell having a cellular characteristic included in the plurality of cellular characteristics for the extracted candidate medium compositions, and a method of searching for and determining candidate medium compositions that will produce a maximum value from the predicted culture results.

[0031] According to the sixteenth aspect, similar to the first and fourteenth aspects, a culture medium panel can be constructed that is small in size and highly compatible with cells with diverse characteristics. The program according to the sixteenth aspect may have a configuration similar to that of the second to fourteenth aspects. Non-transitory tangible recording media (e.g., various magneto-optical recording devices and semiconductor memories) on which computer-readable code for the programs of these aspects is recorded can also be cited as aspects of the present invention. The "non-transitory tangible recording media" mentioned above does not include non-tangible recording media such as carrier signals or propagation signals themselves.

[0032] As described above, the constructing method, constructing device, and constructing program of the present invention make it possible to construct a medium panel that is small in size and highly compatible with cells with diverse characteristics.

[0033] FIG. 1 is a schematic diagram showing how a culture medium panel is constructed. FIG. 2 is a diagram showing the configuration of a culture medium panel construction system. FIG. 3 is a block diagram showing the configuration of a processing unit. FIG. 4 is a block diagram showing the configuration of a culture result prediction unit. FIG. 5 is a diagram showing data stored in a storage unit. FIG. 6 is a flowchart showing a culture medium panel construction method and the processing of the construction program. FIG. 7 is a flowchart showing a culture result prediction method. FIG. 8 is a diagram showing an outline of a simulator for predicting biological behavior quantities. FIG. 9 is a schematic diagram showing an example of a cellular metabolic pathway. FIG. 10 is a diagram showing an outline of the Michaelis-Menten equation. FIG. 11 is a diagram showing an outline of a conventional cell metabolism model. FIG. 12 is a schematic diagram showing how a cell death model is created by machine learning. FIG. 13 is a diagram conceptually showing a method for constructing a culture medium panel. FIG. 14 is a diagram showing a phylogenetic tree of CHO cells. FIG. 15 is a diagram showing how a culture medium panel is divided into subsets. FIG. 16 is a schematic diagram showing the search effect by simulation. FIG. 17 is a table showing the results of a comparison between a conventional method and a method according to an embodiment of the present invention.

[0034] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The following detailed description will be given of embodiments of a configuration method, a configuration device, and a configuration program according to the present invention, with reference to the accompanying drawings as necessary.

[0035] [Processing Overview] The following describes an example of exploring the composition of a medium for antibody-producing CHO cells, with the productivity targeted for optimization set to the antibody production volume. However, the target cells of the present invention may be CHO cells, HEK cells that produce viruses for gene therapy, or iPS cells for regenerative medicine. The target productivity may also be one or more of cell proliferation, antibody production volume or virus production volume, and the success rate of inducing differentiation of target cells into specific tissues. Even when these target cells or productivity are selected, embodiments and examples similar to those described below can be constructed. Furthermore, the culture method may be appropriately selected from well plates (microtiter plates, microplates), flasks, batch, fed-batch, perfusion culture, etc.

[0036] Here, a medium is a mixture of nutrients for culturing cells, generally including, but not limited to, amino acids, vitamins, and metals. Therefore, it can be expressed as a multidimensional vector (or a point in a multidimensional space (hereinafter referred to as "medium composition space") in which each dimension corresponds to a nutrient). In the present invention, the amount and content ratio of part or all of the composition (nutrients) of such a medium can be optimized. Note that the medium composition space is generally multidimensional (the number of dimensions varies depending on the number of nutrients), but in the following explanation, it may be shown as one-dimensional for simplicity.

[0037] Furthermore, for example, "medium" refers to a medium for CHO cells (CHO: Chinese Hamster Ovary), "cells" refers to certain CHO cells (including not only original CHO cells but also various lineages of CHO cells generated by mutation, modification, etc., such as CHO-DG44, CHO-K1, and CHO-S), and "culture results" refers to the amount of antibody produced after culturing the cells for a predetermined period of time. However, as mentioned above, the medium, cells, and culture results of the present invention are not limited to these examples.

[0038] [First embodiment] [Culture medium panel configuration system] Fig. 1 is a diagram showing a schematic diagram of a culture medium panel being configured by a culture medium panel configuration system 10 (configuration device) according to the first embodiment. Fig. 2 is a diagram showing the configuration of the culture medium panel configuration system 10. In the following description, some of the components of the culture medium panel configuration system 10 will be omitted as appropriate.

[0039] The culture medium panel configuration system 10 is a system for configuring a culture medium panel consisting of multiple culture media to optimize cell culture results, and can be implemented using a computer. As shown in FIG. 2, the culture medium panel configuration system 10 includes a processing unit 100 (processor), a memory unit 200, a display unit 300, and an operation unit 400, which are interconnected to transmit and receive necessary information. These components can be installed in various configurations, and each component may be installed in a single location (e.g., within a single cabinet or room) or in separate locations and connected via a network. The culture medium panel configuration system 10 can also connect to an external server 500 and an external database 510 via a network NW, such as the Internet, to acquire information on cell characteristics, culture medium composition, cell lines, etc. as needed, and provide the acquired culture medium panel information to a desired server or database, such as the external server 500 or an external database.

[0040] 3 is a block diagram showing the configuration of the processing unit (processor) 100. The processing unit 100 includes an area setting unit 102, a property enumeration unit 104, a property difference evaluation unit 106, a cell property generation unit 108, a culture result prediction unit 110, a search and enumeration unit 112, a culture medium panel configuration unit 114, an input / output control unit 116, a communication control unit 118, a ROM 130 (Read Only Memory), and a RAM 140 (Random Access Memory).

[0041] 4 is a block diagram showing the configuration of the culture result prediction unit (processor) 110. The culture result prediction unit 110 includes a culture environment input unit 110A, a biological behavior amount prediction unit 110B, a cellular environment change unit 110C, a repeat unit 110D, and an output unit 110E.

[0042] The functions of each unit of the processing unit 100 described above can be realized using various processors. The various processors include, for example, a CPU (Central Processing Unit), which is a general-purpose processor that executes software (programs) to realize various functions. The various processors also include a programmable logic device (PLD), such as an FPGA (Field Programmable Gate Array), whose circuit configuration can be changed after manufacturing. Furthermore, the various processors also include dedicated electrical circuits, such as an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically to execute specific processing.

[0043] The functions of each unit may be realized by a single processor or by a combination of multiple processors. Furthermore, multiple functions may be realized by a single processor. Examples of multiple functions configured by a single processor include: a first configuration, as typified by client and server computers, in which a single processor is configured by combining one or more CPUs and software, and this processor realizes multiple functions; a second configuration, as typified by system-on-chip (SoC), in which a processor is used to realize the functions of the entire system on a single IC (Integrated Circuit) chip; and various functions are thus configured as a hardware structure using one or more of the various processors described above. Furthermore, the hardware structure of these various processors is, more specifically, an electrical circuit (circuitry) that combines circuit elements such as semiconductor devices.

[0044] When the processor or electrical circuit executes software (programs), processor-readable code for the software to be executed is stored in a non-transitory, tangible storage medium such as ROM 130, and the processor references the software. The software stored in the non-transitory, tangible storage medium includes a program (a configuration program for executing the method for configuring a culture medium panel consisting of multiple culture media to optimize cell culture results; hereinafter referred to as the "configuration method") for executing the method of the present invention (a configuration program for executing the method for configuring a culture medium panel consisting of multiple culture media to optimize cell culture results; hereinafter referred to as the "configuration program"). Instead of ROM 130, the code may be recorded in a non-transitory, tangible storage medium such as various optical magnetic recording devices or semiconductor memory. During processing using the software, for example, RAM 140 is used as a temporary storage area, and data stored in, for example, an EEPROM (Electronically Erasable and Programmable Read Only Memory), not shown, may be referenced.

[0045] It should be noted that the above-mentioned "non-transitory and tangible recording medium" does not include non-tangible recording media such as carrier signals or propagation signals themselves.

[0046] The processing using these components of the processing unit 100 will be described in detail later.

[0047] [Configuration of the Storage Unit] The storage unit 200 is composed of various types of non-transitory and tangible storage media such as optical magnetic recording media and semiconductor memory, and its input / output control unit, and stores information such as culture medium compositions, cell characteristics, culture results, etc. In the storage unit 200 of this embodiment, for example, as shown in Fig. 5, a culture medium composition database 210 (e.g., an experimental database), culture medium composition candidates 220, and culture medium panel data 230 are stored in an associated manner.

[0048] [Configuration of Display Unit and Operation Unit] The display unit 300 is equipped with a monitor 310 (display device) and can display input information, information stored in the memory unit 200, results of processing by the processing unit 100, etc. The operation unit 400 includes a keyboard 410 and a mouse 420 as input devices and / or pointing devices, and the user can perform operations necessary to execute the configuration method and configuration program according to the present invention via these devices and the screen of the monitor 310.

[0049] [Configuration Method and Configuration Program Processing] The configuration method and configuration program processing according to the first embodiment will be described below with reference to Figures 1 to 6. Note that, to simplify the description, some elements of the culture medium panel configuration system 10 will be omitted from the illustration as necessary.

[0050] In the first embodiment, antibody-producing CHO cells are targeted, and an example will be described in which a culture medium panel is constructed with the productivity to be optimized being the amount of antibody produced.

[0051] [Area Setting] The area setting unit 102 (processor) sets an area in the medium composition space (step S100; area setting step, area setting process). The "area" corresponds to the search space of the medium composition for constructing the medium panel. The area setting unit 102 may set an area based on the medium composition recorded in the medium composition database 210 shown in Figures 1 and 5, or may set an area based on literature information including catalogs. The area setting unit 102 may limit or expand the composition components to be applied.

[0052] [Enumeration of Cellular Characteristics] The characteristic enumeration unit 104 (processor) enumerates multiple cellular characteristics (step S110; characteristic enumeration step, characteristic enumeration process). The characteristic enumeration unit 104 may enumerate known cellular characteristics based on past experiments or literature information such as papers. The characteristic enumeration unit 104 may also add unknown cellular characteristics (hypothetical cellular characteristics) (for this extension, see the "Enumeration and Extension of Cellular Characteristics" section). While "cellular characteristics" can be defined in various biological or cytological ways, in the present invention, a "cellular characteristic" is defined as "a mechanism that determines the productivity of a cell (such as the cell proliferation rate and antibody production amount described above) when a medium composition is given." Specifically, in this embodiment, a "cellular characteristic" is a mechanism that determines the antibody production amount relative to a medium composition and can be expressed, for example, by a function, a machine learning model, or a mathematical cellular model. Therefore, for example, function coefficients, machine learning parameters, or parameters of a metabolic pathway model in a mathematical cellular model can be considered part of the cellular characteristics in the present invention. Specific examples of cell characteristics will be explained in conjunction with the culture result prediction unit 110 (processor) described below.

[0053] [Prediction of Culture Outcome] The characteristic enumeration unit 104 inputs a region S of the medium composition space (specifically, a medium composition s ∈ region S included in region S) into the culture outcome prediction unit 110 based on the enumerated cellular characteristics (cellular characteristics C; see FIG. 1 ). The culture outcome prediction unit 110 then predicts the culture outcome of a target cell in a medium of a given composition based on the cellular characteristics of the target cell (step S120; prediction step, prediction process). The culture outcome prediction unit 110 may be a function (specifically, a processor, program, memory, etc. that realizes the function of the function), a predictor obtained by machine learning (see FIGS. 8 and 12 ), or a metabolic pathway model simulator (cell simulation). In the first embodiment, such a cell simulation can represent at least one of the prediction means, the prediction function of an unknown cell, and the difference between cellular characteristics in the prediction step or prediction process.

[0054] The culture result prediction unit 110 may construct a model by measuring several points through experiments and, for example, connecting them continuously and smoothly, or may construct a machine learning model with the medium composition as an explanatory variable and the culture result as a target variable, or may construct a metabolic pathway model. Therefore, the cell characteristic C may be, for example, a coefficient of a function serving as a predictor, a machine learning parameter, or a metabolic map in a metabolic pathway model.

[0055] [Culture Result Prediction Method] One aspect of the culture result prediction method performed by the culture result prediction unit 110 described above will be described.

[0056] <<Method for Predicting Cell Cultivation Results>> Figure 7 is a flowchart showing a method for predicting the culture results of target cells using the culture result prediction unit 110 configured as described above. This method includes a culture environment input step of inputting a culture environment consisting of a medium composition and culture conditions for culturing cells, a biological behavior amount prediction step of predicting a biological behavior amount based on the culture environment input in the culture environment input step, and a cellular environment modification step of modifying the cellular environment based on the biological behavior amount predicted by the biological behavior amount prediction step. The method also includes a repeating step of repeating the biological behavior amount prediction step of predicting a biological behavior amount based on the cellular environment modified by the cellular environment modification step and the cellular environment modification step until a predetermined time, which is the culture time, has elapsed. The method also includes an output step of outputting the cellular environment modified by the cellular environment modification step after the predetermined time has elapsed.

[0057] Each step will be described below.

[0058] <<Culture Environment Input Step (Step S12)>> The culture environment input unit 110A performs the culture environment input step (step S12). The culture environment input step is a step of inputting a culture environment consisting of a medium composition and culture conditions for culturing cells. The culture environment is input in response to a user operation and processed by the culture environment input unit 110A.

[0059] The medium composition includes the medium components and the amounts of the medium components in the medium in which the cells are cultured. The culture conditions are set conditions for optimizing the cell culture process, and include the culture method, the size and type of culture vessel, oxygen addition, supplementation of medium and nutrient sources, addition of pH adjusters and carbon dioxide, removal of medium containing growth-inhibitory by-products, and harvesting of the target product.

[0060] Examples of the culture method include whether or not sterilization is performed, whether the culture medium is liquid or solid, the culture temperature, etc. Culture conditions can be set not only to the conditions at the start of culture, but also to include whether or not oxygen is added during the culture period and the amount of oxygen added, whether or not medium and nutrient sources are replenished and the amount of replenishment, addition of a pH adjuster and carbon dioxide, whether or not medium containing growth-inhibiting by-products is removed and the amount of replenishment, and whether or not the target product is harvested and the amount of harvest.

[0061] <<Biological Behavior Amount Prediction Step (Step S14)>> The biological behavior amount prediction unit 110B performs the biological behavior amount prediction step (step S14). The biological behavior amount prediction step is a step of predicting a biological behavior amount based on the culture environment input in the culture environment input step. The biological behavior amount is the change over time in the total number of cells, the change over time in the substances and amounts produced by the cells, and the change over time in the substances and amounts contained in the culture environment. The substances produced by the cells include antibodies produced by the cells, etc. The substances contained in the culture environment include medium components and by-products, etc.

[0062] Fig. 8 is a diagram showing an outline of a simulator for the biological behavior quantity prediction step. As shown in Fig. 8, the biological behavior quantity prediction step is performed using an action mechanism model (biological action mechanism model) and a machine learning model. The action mechanism model uses separate mathematical models for the inside and outside of the cell. For the inside of the cell, the absorption rate F Ａ is used to calculate the cell growth rate F ｇ , and the antibody production rate F ｐ A cell metabolism model is used to calculate the change in the concentration of each component of the culture medium. For the outside of the cell, a culture medium model is used to calculate the change in the concentration of each component of the culture medium. For cell death, a model derived by machine learning is used.

[0063] First, we will explain the cellular metabolism model that operates inside cells. The cellular metabolism model can calculate the cell growth rate and antibody production rate by using a cell culture simulation method that reproduces the mechanism and bioprocess of the cultured cells. The cell culture simulation method can be performed using a modeling approach that includes metabolic flux analysis (MFA) using a genome-scale metabolic model or flux balance analysis (FBA).

[0064] "Flux" or "metabolic flux" refers to the rate at which molecules pass through a target pathway or reaction. Factors that control flux include the catalytic rate of the enzyme in the pathway, the availability (durability) of the substrate, the concentration of the enzyme in the cell, and the proximity of the enzyme in the pathway. "Metabolic flux analysis" is a method for determining the amount of molecules moving based on these factors. "Flux balance analysis" is an analytical method that focuses on stoichiometry and metabolic flow. Even when metabolic constants cannot be fully measured, it analyzes the behavior range and characteristics of the target metabolic circuit based on the structure of metabolic reactions and basic constraints such as the law of conservation of mass.

[0065] Next, intracellular metabolism will be explained using Figure 9. Figure 9 is a schematic diagram showing an example of a metabolic pathway of a cell. For the cell metabolism model, first, the substrates used for metabolism in the culture medium and the uptake constraint conditions, which are the upper and lower limits for uptake of nutrients, etc., are obtained. In Figure 9, substances produced by metabolism are indicated by A, B, C, .... １ , F ２ , F ３ , ... are functions that indicate the time change in the concentration of each substance, for example, F １ is the flux of substance A taken up, and F ２ is the flux of substance B metabolized.

[0066] In the example shown in Figure 9, substance A is converted to substance B in equal amounts due to the law of conservation of mass. Similarly, the amount of substance B is equal to the sum of the amounts converted to substances C and E. In this way, the amount of substance converted is determined by the amount of substrate and nutrients initially taken up by the cells, etc. In addition, reaction rate is a parameter that limits the conversion of substances. For example, as described above, substance A is converted to substance B in equal amounts, but due to reaction rate limitations in a certain unit time, there is an upper limit to the conversion to substance B, and it may be that not all of substance A is converted to substance B. As an uptake constraint, a parameter that limits the conversion of this substance is obtained, and this uptake constraint is obtained using an equation that mathematically models the mechanism by which cells take up medium components, and is applied when calculating the culture results.

[0067] As a mathematical model for calculating this reaction rate, for example, the Michaelis-Menten equation can be used. A schematic diagram of the Michaelis-Menten equation is shown in FIG. 10. The Michaelis-Menten equation is an equation relating to the enzyme reaction rate V. When the substrate concentration S is low, the reaction rate V is proportional to the substrate concentration S, and when the substrate concentration S is high, the reaction rate V converges to a maximum rate Vmax regardless of the substrate concentration S. Another mathematical model that can be used is Fick's law. Fick's law is an equation for calculating the flux, which is the amount passing through a unit area per unit time, and is proportional to the diffusion coefficient D and the substrate concentration gradient on both sides of the membrane. In this embodiment, by using the Michaelis-Menten equation or Fick's law as a constraint condition for taking in the medium, simulations can be performed without calculating conditions for a reaction rate that is faster than the reaction rate and flux calculated by the mathematical model but is within the range of the conversion amount according to the law of conservation of mass. This allows for more accurate culture predictions.

[0068] The simulation calculations may use multiple mathematical models that mimic the state of cells in parallel. For example, a mathematical model for cell proliferation and a mathematical model for bioproduction without cell proliferation can be used in parallel, and the ratio between the multiple models can be changed depending on the culture state, such as the medium concentration.

[0069] Next, we will explain the culture medium model that operates outside the cell. The culture medium model is a model that calculates the concentration changes in the culture medium surrounding the cell when metabolism occurs under the conditions of the above-mentioned cell metabolic model, and a cell signaling model expressed as a differential equation can be used. The concentration changes of each component can be calculated by solving ordinary differential equations over time t using the Runge-Kutta method.

[0070] Here, a conventional cell metabolism model including a cell death model will be described. FIG. 11 is a diagram showing an outline of the conventional cell metabolism model. The main cell culture activities over the culture period can be divided into cell proliferation, antibody production, and cell death (cessation of activity). As shown in FIG. 11 , these activities include a proliferation phase in the early stage of culture, in which cells proliferate; an antibody production phase in the middle stage, in which the cell proliferation rate decreases and the antibody production rate increases; and a cell death phase in which cell activity ceases and cell death occurs. Conventional culture prediction simulations use a cell metabolism model to calculate these three culture activities, with a proliferation model for cell proliferation, a production model for antibody production, and a cell death model for cell death. However, the cell death model was implemented using a simplified model with constants, which did not achieve sufficient accuracy. Therefore, in this embodiment, as shown in FIG. 8 , a simulation is performed using a trained model (one aspect of a predictor) derived from machine learning as a cell death model, thereby calculating culture prediction results with high accuracy.

[0071] FIG. 12 is a schematic diagram of creating a cell death model by machine learning. A random forest method can be used as a method for creating a trained model by machine learning. When constructing a cell death model by machine learning, the number of live cells, the concentrations of 20 types of amino acids, and the ammonia concentration are input as input data. Furthermore, the cell viability is input as output data. The data used is obtained by actually culturing cells, and, for example, 80% of the measured data is used for learning, and 20% is used for testing. This allows the creation of a cell death model (one aspect of a predictor).

[0072] Note that the creation of trained models using machine learning is not limited to cell death models. For data items that are difficult to model based on mechanisms, such as cell death, creating trained models using machine learning allows for more accurate calculation of culture prediction results. Examples of such data items include growth inhibition and antibody production suppression. When creating a growth inhibition model using machine learning, the cell growth amount (growth ratio) can be used as the output data. When creating an antibody production suppression model, the antibody production amount (production ratio) can be used as the output data.

[0073] In the biological behavior quantity prediction step of this embodiment, a trained model of cell death created by machine learning is used, which allows the biological behavior quantity to be measured with higher accuracy, and therefore allows the predicted results of culture to be calculated with high accuracy.

[0074] The biological behavior quantity prediction step (step S14) can be performed using a CHO cell simulator (a device or system that simulates the biological behavior of CHO cells using a mechanism of action model and a trained model of cell death created by machine learning) that includes these models, based on the culture conditions (medium component concentrations) and uptake constraint conditions acquired in the culture condition input step (step S12).

[0075] <Cellular Environment Changing Step (Step S16)> The cellular environment changing unit 110C performs the cellular environment changing step (Step S16). The cellular environment changing step changes the cellular environment based on the biological behavior amount predicted by the biological behavior amount prediction step. The cellular environment refers to the total number of cells and the substances and amounts produced by the cells.

[0076] The biological behavior amount prediction step calculates the change over time in the number of cells after the culture time t has elapsed, the change over time in the substances and amounts produced by the cells, and the change over time in the substances and amounts contained in the culture environment. The cellular environment change step reflects the change in the number of cells after the culture time t has elapsed and the changes in the components in the culture medium in the cellular environment.

[0077] <<Repetition Step>> The repetition unit 110D controls the execution of the repetition step, which is a step of repeating the biological behavior amount prediction step (step S14) and the cellular environment modification step (step S16).

[0078] After the cell environment changing step (step S16) is completed, the repeating unit 110D determines whether a predetermined culture time has elapsed (step S18). If the predetermined culture time (e.g., 14 days) for the calculation has elapsed (if YES), the calculation is terminated and the process proceeds to the output step (step S24). If the predetermined culture time has not elapsed (if NO), the process determines whether to replenish the culture medium, nutrients, etc. (step S20). If the culture medium, nutrients, etc. will not be replenished (if NO), the process returns to the biological behavior quantity prediction step (step S14). If the culture medium, nutrients, etc. will be replenished (if YES), the culture medium concentration changing step (step S22) is performed.

[0079] If the culture environment input step (step S12) specifies that the culture medium and nutrients be replenished after a predetermined time has elapsed, a culture medium concentration change step (step S22) is performed. The replenishment of the culture medium and nutrients includes the replenishment of a pH adjuster and carbon dioxide in addition to the culture medium and nutrients. The culture medium concentration change step changes the culture medium concentration by adding the replenished culture medium and nutrients to the cellular environment at the time the cellular environment change step is completed.

[0080] Furthermore, the medium concentration changing step can not only replenish the medium and nutrients, etc., but also remove the medium containing growth-inhibiting by-products and change the medium concentration by harvesting the target product. If the removal of the medium containing growth-inhibiting by-products and the harvesting of the target product are set as culture conditions in the culture environment input step (step S12), the medium concentration is changed by removing the removed growth-inhibiting by-products and the harvested target product from the cellular environment at the time the cellular environment changing step is completed.

[0081] When the repeating step is performed, the uptake constraint conditions are acquired based on the cellular environment changed in the cellular environment changing step (step S16) or the medium concentration changed in the medium concentration changing step (step S22), and the biological behavior amount prediction step (step S14) and the cellular environment changing step (step S16) are performed. Thereafter, the biological behavior amount prediction step (step S14) and the cellular environment changing step (step S16) are repeated until a predetermined culture time has elapsed (step S18).

[0082] After the predetermined culture period has elapsed (YES in step S18), the calculation is terminated. As a result, the number of cells and the amount of bioproduction or ammonia produced during the culture period can be determined. Furthermore, by plotting the amount of biological behavior at each point in time during the repeated process, it is possible to determine the cell growth curve and the progress of antibody production.

[0083] <<Output Step (Step S24)>> The output unit 110E performs the output step (step S24), which is a step of outputting the cellular environment changed in the cellular environment changing step (step S16).

[0084] The results output by the output step can include the number of cells grown during the culture period, and the amount of bioproduction produced, or the amount of bioproduction of antibodies such as ammonia. Also, a cell growth curve, progress of antibody production, etc. can be output. The results to be output may be the results after a predetermined time has elapsed, or the results may be output after any period of time has elapsed during the culture.

[0085] [Search and Enumeration of Medium Compositions] Returning to Figures 1 to 6, the explanation continues. The search and enumeration unit 112 (processor) uses the culture result prediction unit 110 (predictor) to extract candidate medium compositions for which the cell culture results for various medium compositions s within the region S are maximized. Based on the predicted culture results for the extracted candidate medium compositions, the search and enumeration unit 112 searches for and enumerates each of the cellular characteristics included in the above-described multiple cellular characteristics (step S130; search and enumeration step, search and enumeration process). The search and enumeration unit 112 may search for a maximum value as a search problem for function optimization. For example, if the gradient of the function serving as the predictor can be defined, the search and enumeration unit 112 may search for a maximum value using Newton's method or its extension method, or may select the maximum value from a list of random values ​​of the function serving as the predictor. The search and enumeration unit 112 can store the obtained candidate medium compositions (candidate R) in the memory unit 200 as candidate medium compositions 220.

[0086] As is generally known, if the gradient can be defined or approximately calculated, the search and enumeration unit 112 can calculate a convex point in the set region where the gradient is zero as a "maximum." Even if the gradient is not differentiable, it can be approximately calculated from the values ​​of two nearby points by calculating the difference between the values ​​of adjacent points. Alternatively, the search and enumeration unit 112 may randomly calculate values ​​and then extract a subregion from the randomly calculated values ​​to calculate the maximum. Furthermore, other optimization methods, such as the Nelder-Mead method, which do not use derivatives, may also be used. The search and enumeration unit 112 can also apply various other optimization methods for linear or nonlinear functions to calculate the maximum.

[0087] [Configuration of Culture Medium Panel] The culture medium panel configuration unit 114 (processor) configures a culture medium panel from the candidate culture medium composition R obtained in step S130 (step S140; configuration step, configuration process). Details of the culture medium panel configuration will be described later.

[0088] [Presenting the Culture Medium Panel] The input / output control unit 116 (processor) presents the obtained culture medium panel information (step S150 in FIG. 6; information presentation process, information presentation processing). The input / output control unit 116 may store the culture medium panel information in the storage unit 200 as culture medium panel data 230 (see FIG. 5), or may provide the information to an external server 500, an external database, etc. via the communication control unit 118 and the network NW.

[0089] [Enumerating and Expanding Cellular Characteristics] The characteristic enumeration unit 104 may generate and add unknown cellular characteristics (virtual cellular characteristics) when enumerating cellular characteristics. That is, a cellular characteristic can be generated by the cellular characteristic generation unit 108 (characteristic generator) based on a difference constraint L (a constraint on the difference) from known cellular characteristics (see FIG. 1 ). For example, if the cellular characteristic C is defined as a coefficient of a function F (one embodiment of the culture result prediction unit 110), the difference constraint L may be the variation range of the coefficient. If the cellular characteristic C is defined as a machine learning parameter, the difference constraint L may be the variation range of the parameter. Alternatively, the difference constraint L may be a difference between models, such as the Hamming distance between network matrix representations of metabolic pathway models.

[0090] Since these can be measured between known cell characteristics, the characteristic difference between the unknown cell and the known cell can be set with reference to the measured difference. That is, the difference constraint L may be based on literature information, or the characteristic difference between known cell characteristics may be measured (evaluation step, evaluation process for evaluating the characteristic difference between cells), and the cell characteristic generation unit 108 (characteristic generator) may set the difference constraint L (e.g., equal or a constant multiple) based on the measured difference constraint L. Based on the set difference constraint L, the cell characteristic generation unit 108 can generate a characteristic (virtual cell characteristic) of an unknown cell whose difference from the cell characteristic of a known cell group satisfies the difference constraint L (constraint condition) while controlling the characteristic of the unknown cell (generation step, generation process). When enumerating multiple cell characteristics in the characteristic enumeration step or characteristic enumeration process, the characteristic enumeration unit 104 can include one or more unknown cell characteristics (virtual cell characteristics) generated in this manner.

[0091] [Details of Culture Medium Panel Configuration] The configuration of the culture medium panel (i.e., configuring culture medium panel P from candidate R in FIG. 1) is explained with reference to FIG. 13. In FIG. 13, the horizontal axis represents the culture medium composition, and the vertical axis represents the amount of antibody production. The solid line represents the amount of antibody production for known cells with various culture medium compositions, and is the result of predicting the culture result based on the known cell characteristics using a predictor such as a function (culture result prediction unit 110). Here, one line is shown, but there may be multiple lines. On the other hand, the dotted line represents the amount of antibody production for unknown cells, and is also the result of predicting the culture result based on the unknown cell characteristics using a predictor.

[0092] Here, the culture medium panel configuration unit 114 (processor) selects the culture medium to be added to the culture medium panel based on the shapes of the solid and dotted lines. The maximum value or its vicinity is a strong candidate, and points close to the maximum value for multiple cells are particularly suitable. For example, points indicated by an upward-pointing triangle △ (Culture medium 1, Culture medium 2A, Culture medium 2B) are suitable examples, while points indicated by a downward-pointing triangle ▽ (unsuitable) are unsuitable examples.

[0093] Furthermore, selecting a variety of media may be redundant, potentially enabling the selection of unknown cells with slightly different characteristics. For example, if two were to be selected from the three candidates mentioned above, medium 1 and medium 2B would be desirable, since it would be preferable to select only medium 2B from mediums 2A and 2B, which have similar compositions. Furthermore, medium 1 is expected to produce a slightly lower amount of antibodies than mediums 2A and 2B for known cells, but is particularly favorable for unknown cell A. Therefore, it is desirable for the medium panel construction unit 114 to select a medium from such candidates.

[0094] Abstractly, a diverse culture medium can be described as a "selection method that broadens the distribution of candidate culture medium compositions in the culture medium composition space." The culture medium panel construction unit 114 may select, for example, candidate culture mediums that narrow the correlation between candidate culture mediums, that broaden the distance between candidate culture mediums, or that increase diversity as defined by information gain when constructing the culture medium panel. Adding a culture medium to the culture medium panel can provide information on the cell characteristics corresponding to the added medium (culture results in that medium; specifically, antibody production levels, for example). Therefore, adding a culture medium to the culture medium panel can be considered a virtual "observation" of cell characteristics. In this case, a culture medium whose addition is expected to significantly reduce the uncertainty of the cell characteristics of unknown cells is a "culture medium with high information gain," while a culture medium whose addition is expected to result in little reduction in uncertainty is a "culture medium with low information gain." A culture medium with high information gain is a culture medium with high diversity, and the culture medium panel construction unit 114 preferably prioritizes adding such a culture medium to the culture medium panel. In this way, by using information gain, it is possible to ensure the diversity of culture media by taking into account the uncertainty of functions depending on cell characteristics.

[0095] The culture medium panel constructing unit 114 can calculate correlations and distances between culture medium compositions by directly comparing the culture medium compositions with each other.

[0096] In the example of Figure 13, the cell characteristics are thought to be relatively similar to each other, but since there may be new compositions, such as medium 1, that are particularly suitable for novel cells, it can be said that selecting media with diverse compositions is particularly effective. That is, for example, if the unknown cell X actually obtained has characteristics similar to A, good culture results can be expected with medium 1, and if it has characteristics similar to B, good culture results can be expected with medium 2B. In this way, the present invention makes it possible to improve the compatibility of media with diverse characteristics by constructing a medium panel (set) consisting of multiple media, rather than necessarily using a single medium.

[0097] The culture medium panel construction unit 114 may weight candidate culture medium compositions based on the frequency of appearance or probability of the cell characteristics, and construct a culture medium panel based on the weighting results. For example, in the example of Figure 13, if it is assumed that a cell similar to unknown cell B has a high frequency of appearance and if additional culture media can be added to the culture medium panel (if there is room for more culture media), the culture medium panel construction unit 114 may select both culture media 2A and 2B. The culture medium panel construction unit 114 may arbitrarily assign weights based on the frequency of appearance or probability, or may consider unknown characteristics to have a lower frequency of appearance or probability (lower weight) depending on the difference from the known characteristics.

[0098] An example of constructing a culture medium panel based on the frequency of occurrence of cell characteristics will be described below. For example, if a certain culture medium panel is applied to existing cells 10 times and finds that it matches cell characteristic A 7 times and cell characteristic B 3 times, then the frequencies (or probabilities) of occurrence are A = 0.7 and B = 0.3. Therefore, it is conceivable to adopt a larger proportion of the culture medium composition that maximizes the culture result for cell characteristic A in the culture medium panel.

[0099] The culture medium panel construction unit 114 may use, as the frequency of occurrence of cell characteristics, information actually given by interviewing a customer (a person who receives a culture medium panel constructed according to the present invention and cultures cells), information determined based on performance records (for example, the closer the cell is to past customer cells, the higher the frequency of occurrence), or information determined by simulation (for example, the more frequently a characteristic is generated in the process of refining virtual cell characteristics, the "high frequency of occurrence" is determined).

[0100] [Expansion of Culture Medium Panel Configuration: Utilization of Cell Lineages] Next, additional configuration modes of the culture medium panel will be described with reference to Figures 14 and 15. Even in the case of CHO cells, for example, it is known that various CHO cell lines are not "equally spaced" from one another, and that relationships like a phylogenetic tree can be modeled based on derivation relationships, etc. Specifically, as shown in Figure 14, there are multiple CHO cell lines with different characteristics, such as "CHO-K1," "CHO-S," and "CHO-DG44," in addition to the original CHO cell (CHO-ori).

[0101] The phylogenetic tree in FIG. 14 is described in Non-Patent Document 1 below.

[0102] [Non-Patent Document 1] "Wurm, FM; Wurm, MJ Cloning of CHO Cells, Productivity and Genetic Stability A Discussion. Processes 2017, 5, 20." [Retrieved March 1, 2022], Internet (https: / / www.researchgate.net / publication / 316314849_Cloning_of_CHO_Cells_Productivity_and_Genetic_Stability-A_Discussion / link / 59410f0045851566e1beae32 / download) The above paper is cited in accordance with http: / / creativecommons.org / licenses / by / 4.0 / .

[0103] The above-described phylogenetic tree-like relationship is not limited to CHO cells; HEK cells also have phylogenetic trees, and such relationships may also emerge with iPS cells through long-term culture. In the present invention, "cells" generally refer to host cells containing a certain gene and antibody-producing cells (sometimes referred to as "cell lines") established in this manner, into which an antibody-producing gene has been inserted. Such lineages or cell lines may be defined by genes or by reference to commonly known lineages (e.g., "CHO-K1," "CHO-S," "CHO-DG44," etc., in the case of CHO cells). While various definitions of "cell line" exist, it can be defined, for example, as "cells isolated from a living organism or cells that have been modified in some way, such as genetically, so that they can be stably grown and cultured for a long period of time while maintaining certain properties."

[0104] The culture medium panel construction unit 114 may utilize this relationship to divide the culture medium panel into subsets (small subsets of the culture medium panel may be constructed within the culture medium panel; Figure 15). For example, if the culture medium panel were to accommodate all possible cell types (all DG44 / K1 / S lineages in the example of Figure 15), the size of the culture medium panel would be large or the medium composition would be too coarse (Universal Panel). This "Universal Panel" is an example of a "subset not dependent on a specific cell lineage." Therefore, if the lineage of an unknown cell can be known, it is possible to configure a panel with a tighter correspondence while reducing the size of the culture medium panel (Specific Panel) by associating at least one known cell lineage with one of the subsets (for example, by limiting one of the subsets to the known cell lineage S). This "Specific Panel" is an example of a "subset dependent on a specific cell lineage."

[0105] In this way, when an unknown cell is actually given, the culture medium panel construction unit 114 may choose between a Universal Panel and a Specific Panel to apply based on prior knowledge of the cell's characteristics (for example, if only the lineage is known, it may provide the corresponding Specific Panel, and if the characteristics including the lineage are unknown, it may provide a Universal Panel).

[0106] Furthermore, when the lineage of the unknown cell cannot be determined, the culture medium panel construction unit 114 may divide the culture medium panel into subsets and include a subset that is not dependent on a specific cell lineage (the Universal Panel described above) in the subsets that constitute the culture medium panel. For example, when determining three culture media for a culture medium panel, if there are three cell lineages, the culture medium panel construction unit 114 will use media compatible with representative cells of each lineage in the "Universal Panel." This is expected to prevent significant loss of compatibility regardless of the cell lineage being targeted. On the other hand, when constructing a culture medium panel for a subset dependent on a specific cell lineage (Specific Panel), for example, cell line S (a lineage of CHO cells; see FIG. 14), the culture medium panel construction unit 114 will use media compatible with the representative cell S of lineage S and cells S1 and S2 derived from the representative cell S. This may result in low compatibility when targeting unexpected cell lineages (e.g., cell lineage K, etc.), but since this medium composition only needs to be adopted when the cell lineage is known to be S, better culture results can be expected for cells with characteristics similar to those of cells S1 and S2 than for a medium panel composed only of medium compatible with cell lineage S.

[0107] [Application of Submodular Optimization Method] The culture medium panel configuration unit 114 may further configure the culture medium panel by applying a submodular optimization method. "Submodular optimization" is an optimization method that utilizes the submodularity of a set function, and "submodularity" is a concept that corresponds to convexity or concavity (in a continuous function) defined for a set, and is known to be applicable to combinatorial optimization.

[0108] Specifically, the maximum antibody production amount P(M) for cells with medium panel M and cell characteristics C is defined by the following formula (1): P(M)=max{F_C(m∈M)} (1)

[0109] The medium panel M can be any element in the medium composition space S, but as mentioned above, the medium composition space S may have elements of the maximum value or its vicinity predicted by the predictor F (culture result prediction unit 110). In this case, the object of the present invention is to maximize the maximum antibody production amount P(M) when an unknown cell characteristic C is given, and the following equation (2) holds for any medium k. [Equation 2] P(M∪{k})=max{P(M),F_C(k)} ... (2)

[0110] Therefore, when another culture medium panel N⊂M is considered, the following equations (3) and (4) hold: [Equation 3] P(N∪{k})-P(N)=max{0,F_C(k)-P(N)}...(3) [Equation 4] P(M∪{k})-P(M)=max{0,F_C(k)-P(M)}...(4)

[0111] Moreover, since P(M) ≥ P(N) is clear, the following equation (5) holds: [Equation 5] P(N∪{k})-P(N) ≥ P(M∪{k})-P(M) ... (5)

[0112] Therefore, the maximum antibody production P(M) is a submodular function. When the size of the culture medium panel is t, the problem of constructing the optimal culture medium panel for a given cell characteristic can be expressed as the following equation (6): [Equation 6] argmax_M{P(M)}, |M|≦t ... (6)

[0113] This problem is an NP-Hard problem, but an approximate solution method is known, for example, from Non-Patent Document 2 below.

[0114] [Non-Patent Document 2] Nemhauser, George L., Laurence A. Wolsey, and Marshall L. Fisher. "An analysis of approximations for maximizing submodular set functions-I." Mathematical Programming 14.1 (1978): 265-294. [Retrieved March 1, 2022], Internet (https: / / www.cs.toronto.edu / ~eidan / papers / submod-max.pdf)

[0115] This invention deals with the case where multiple cell characteristics may exist. This can also be considered as a problem of improving as much as possible the "worst case result" for all cell characteristics to be considered in the constructed culture medium panel. In this case, when the set of cell characteristics C to be considered using the method of the present invention is defined as C', the submodular optimization method can be extended and defined as a problem of maximizing the minimum value among multiple submodular functions, as shown in the following equation (7): [Equation 7] argmax_M{min_(C∈C'){P(M,C)=max{F_C(m∈M)}}}, |M|≦t (7)

[0116] For this case as well, an approximate solution method is known as an extension of a submodular function, for example, from the following Non-Patent Document 3, and such a solution method may be applied.

[0117] [Non-Patent Document 3] Krause, Andreas, et al. "Robust Submodular Observation Selection." Journal of Machine Learning Research 9.12 (2008). [Retrieved March 1, 2022], Internet (https: / / jmlr.csail.mit.edu / papers / volume9 / krause08b / krause08b.pdf)

[0118] In other words, one aspect of the present invention applies a submodular optimization method to assign specific components such as cells, cell characteristics, and medium composition, and also proposes a problem-solving method that introduces domain-specific elements.

[0119] [Application of Simulation Model] As described above, a cell simulator or a function similar thereto may be applied to the function corresponding to the predictor F (one aspect of the culture result prediction unit 110) and the cell characteristic C.

[0120] [Example] The search effect by simulation is shown. A cell simulator (predictor F; culture result prediction unit 110) was constructed using the method described in the section "Culture result prediction method," and unknown and known cells were reproduced by observing approximate function shapes. The general trend is shown in a schematic diagram in Figure 16. The solid line shows the relationship between medium composition and antibody production amount for known cells (so-called predictor F). The dotted line corresponds to possible unknown cells, and the dashed line corresponds to impossible unknown cells. (It is assumed that unknown cells have a characteristic that their medium composition, which takes an extreme value, is close to that of one of the known cells.)

[0121] The results of comparing the following four methods are shown in Figure 17. (Method 1A) A method in which medium compositions are evenly distributed based on the conventional method ("Even A" in the figure) (Method 1B) A method in which medium compositions are evenly distributed based on the conventional method ("Even B" in the figure), but the size of the medium panel is larger than the other methods (the number of media included in the medium panel is two more than the other methods) (Method 2) A method in which medium compositions are arranged in order at the maximum values ​​of known characteristics and at a certain distance from each other ("Known extreme values" in the figure; one embodiment of the present invention) (Method 3) A method in which medium compositions are not only matched to the maximum values ​​of known characteristics, but also take into account the characteristics of unknown cells, and in which compositions are not simply arranged far apart but are arranged densely in areas with high value variation ("Unknown prediction" in the figure; one embodiment of the present invention)

[0122] As a result, at least one of Methods 2 and 3, which are embodiments of the present invention, produced the best results, and Method 1 alone never produced the best results. In particular, Method 2 was always the best for known cells, and Method 3 was always the best for possible unknown cells. Specifically, the results were as follows: <Method 1A: Uniform> Evenly distributing the medium compositions may coincidentally correspond to an unlikely unknown cell, but generally, a medium far from the medium composition that maximizes the results for known cells (antibody production; an example of culture results) may be "wastefully" selected, and evenly distributing the medium in high-dimensional media is not possible in the first place. <Method 1B: Uniform> Although the set range and spacing of the medium compositions improved the results for some cell characteristics (e.g., Fb2), the results for Fa5 were worse than those for Method 1A, and evenly distributing Fa1 and Fb1 showed almost no suitability. <Method 2: Known extreme values> (One embodiment of the present invention) By arranging the medium compositions at the maximum values ​​of the known properties and at a certain distance from each other, it was shown that it is possible to deal with possible unknown properties to a certain extent. <Method 3: Known extreme values> (One embodiment of the present invention) By not only making the medium compositions correspond to the maximum values ​​of the known properties but also taking into account the properties of the unknown cells, and by not simply arranging the compositions at a distance from each other but also arranging them closely in areas with high value variation, it was shown that it is possible to deal with unknown properties well.

[0123] Thus, it was confirmed that, according to the embodiment of the present invention that takes into account the maximum value of the cell culture results, it is possible to construct a medium panel that is small in size and optimal for cells with diverse characteristics.

[0124] [Effects of the First Embodiment] As described above, according to the culture medium panel configuration system 10 (configuration method, configuration device, configuration program) of the first embodiment, it is possible to configure a culture medium panel that is highly compatible with cells of various characteristics, and further, it is possible to configure a culture medium panel that is highly compatible with small-sized cells whose characteristics are unknown.

[0125] Although the embodiment and other examples of the present invention have been described above, the present invention is not limited to the above-described aspects and various modifications are possible.

[0126] 1 Culture medium 2A Culture medium 2B Culture medium 10 Culture medium panel construction system 100 Processing unit 102 Area setting unit 104 Property enumeration unit 106 Property difference evaluation unit 108 Cell property generation unit 110 Culture result prediction unit 110A Culture environment input unit 110B Biological behavior quantity prediction unit 110C Cell environment change unit 110D Repeat unit 110E Output unit 112 Search enumeration unit 114 Culture medium panel construction unit 116 Input / output control unit 118 Communication control unit 200 Memory unit 210 Culture medium composition database 220 Culture medium composition candidate 230 Culture medium panel data 300 Display unit 310 Monitor 400 Operation unit 410 Keyboard 420 Mouse 500 External server 510 External database S12 to S24 Each step of the culture result prediction method S100 to S150: Steps of the culture medium panel construction method

Claims

1. A method for constructing a culture medium panel consisting of a plurality of culture media that optimizes cell culture results, comprising: A region setting step of setting a region of a medium composition space; a characteristic enumeration step of enumerating a plurality of cell characteristics; A prediction step of predicting a culture result of a target cell in a given medium composition based on a cellular characteristic of the target cell; a searching and enumerating step of searching for and enumerating medium compositions that maximize the cell culture result for each of the plurality of cell characteristics; A step of constructing the medium panel from candidates of medium compositions corresponding to the maximum values; by a processor; The processor, in the search and enumeration step, Extracting candidate medium compositions from the region of the medium composition space; predicting a culture result of a cell having a cellular characteristic included in the plurality of cellular characteristics for the extracted candidate medium composition by the prediction step; Searching and determining candidates for the medium composition that give the maximum value from the predicted culture results; How to configure.

2. The processor, in the configuring step, The method of claim 1 , wherein the medium panel is constructed such that the distribution of the candidate medium compositions in the medium composition space is broadened.

3. The processor, in the configuring step, The method of claim 2, wherein the culture medium panel is configured such that the correlation between the candidates for culture medium composition in the culture medium composition space is reduced, the distance between the candidates for culture medium composition in the culture medium composition space is increased, or the diversity defined by the information gain of the candidates for culture medium composition in the culture medium composition space is increased.

4. The processor, in the configuring step, The method according to any one of claims 1 to 3, further comprising weighting the candidates for medium composition based on the frequency of occurrence of a cell characteristic, and configuring the medium panel based on the weighting result.

5. The processor, An evaluation step of evaluating characteristic differences between cells; a generating step of generating a virtual cell characteristic, the difference of which between the virtual cell characteristic and a cell characteristic of a known cell group satisfies a constraint condition, while controlling the virtual cell characteristic based on the evaluation; Run When enumerating the plurality of cell characteristics in the characteristic enumeration step, one or more of the generated virtual cell characteristics are included. A method according to any one of claims 1 to 3.

6. The processor, in the generating step, The configuration method according to claim 5 , wherein the constraint condition is set based on characteristic differences between cells in the known cell group.

7. The method of any one of claims 1 to 3, wherein the processor divides the culture medium panel into subsets and corresponds at least one known cell lineage to one of the subsets.

8. 8. The method of claim 7, wherein the processor divides the media panel into subsets, and includes in the subsets a subset that is not dependent on a particular cell lineage.

9. The method of claim 1 , wherein the processor configures the culture medium panel using a sub-modular optimization technique during the configuration step.

10. The configuration method according to any one of claims 1 to 3, wherein the processor represents at least one of the means of prediction in the prediction step, a prediction function of an unknown cell, and a difference between cell characteristics by a cell simulation including a metabolic pathway model.

11. The processor, in the prediction step, predicting the amount of biological behavior of the target cell based on the given medium composition and culture conditions; modifying a cellular environment of the target cell based on the predicted biological behavior amount; repeating the prediction of the biological behavior quantity and the modification of the cellular environment based on the modified cellular environment; The method according to any one of claims 1 to 3, further comprising the step of predicting the culture outcome by:

12. The processor, The prediction is performed using the biological behavior quantities of the change over time in the total number of cells, the change over time in the amount of a substance produced by the target cells and the change over time in the amount of the substance, and the change over time in the amount of a substance contained in the culture environment of the target cells, The method of claim 11 , wherein the change is made to the total number of the target cells, and the substances and amounts of the substances produced by the target cells as the cellular environment.

13. The method according to claim 1 , wherein the target cell is any one of a CHO cell, a HEK cell, and an iPS cell.

14. The method according to any one of claims 1 to 3, wherein the culture results are one or more of the amount of antibody or virus produced by the target cells, the proliferation of the target cells, and the success rate of inducing differentiation of the target cells into a specific tissue.

15. An apparatus for constructing a culture medium panel including a plurality of culture media for optimizing a cell culture result, the apparatus comprising: a processor; The processor, A region setting process for setting a region in a medium composition space; a characteristic enumeration process for enumerating a plurality of cell characteristics; A prediction process for predicting a culture result of a target cell in a given medium composition based on a cellular characteristic of the target cell; a searching and enumerating process for searching and enumerating medium compositions that maximize the cell culture results for each of the plurality of cell characteristics; A process of constructing the medium panel from candidates of medium compositions corresponding to the maximum values; Run In the search and enumeration process, Extracting candidate medium compositions from the region of the medium composition space; predicting a culture result of a cell having a cellular characteristic included in the plurality of cellular characteristics for the extracted candidate medium composition by the prediction process; Searching and determining candidates for the medium composition that give the maximum value from the predicted culture results; Component device.

16. A configuration program that causes a processor to execute a method for configuring a culture medium panel consisting of a plurality of culture media that optimizes a cell culture result, The method comprises: A region setting step of setting a region of a medium composition space; a characteristic enumeration step of enumerating a plurality of cell characteristics; A prediction step of predicting a culture result of a target cell in a given medium composition based on a cellular characteristic of the target cell; a searching and enumerating step of searching for and enumerating medium compositions that maximize the cell culture result for each of the plurality of cell characteristics; A step of constructing the medium panel from candidates of medium compositions corresponding to the maximum values; having In the search and enumeration step, Extracting candidate medium compositions from the region of the medium composition space; predicting a culture result of a cell having a cellular characteristic included in the plurality of cellular characteristics for the extracted candidate medium composition by the prediction step; A method for searching for and determining a candidate medium composition that gives the maximum value from the predicted culture result. Configuration program.

17. A non-transitory computer-readable recording medium having the program according to claim 16 recorded thereon.

Images