Secretion product evaluation method and secretion product evaluation system

Abstract

Provided is a technique capable of evaluating a property of a secretion product with high accuracy.　Specifically provided are a secretion product evaluation method and others, the secretion product evaluation method comprising: a positional information identification step for identifying positional information of at least one secretory cell and at least one antigen-holding particle on the basis of optical information relating to a carrier that holds the secretory cell and the antigen-holding particle; a secretion product information acquisition step for acquiring information relating to a secretion product that binds to a binding target molecule held by the secretory cell and the antigen-holding particle; and a secretion product evaluation step for evaluating a property of the secretion product on the basis of the positional information and the information relating to the secretion product.

Description

Secretion evaluation method and secretion evaluation system

[0001] This technology relates to a secretion evaluation method and a secretion evaluation system. More specifically, this technology relates to a secretion evaluation method and a secretion evaluation system for evaluating the characteristics of a secretion that binds to a bound molecule carried by an antigen-carrying particle.

[0002] In order to analyze the reactivity of cells, a technology for evaluating the characteristics of secretions is required. Here, Non-Patent Document 1 discloses, for example, a method of evaluating the amount of an antibody showing specific binding by measuring a region excluding the cell region from the entire region of a carrier holding cells.

[0003] Also, for example, Patent Document 1 discloses a preparation step of preparing a population of biological particles containing biological particles to which a first capture substance for capturing a secreted substance is bound, a first capture step of binding the secreted substance generated by placing the population of biological particles under predetermined conditions to the first capture substance, and a second capture step of binding the secreted substance bound to the first capture substance to a second capture substance for capturing a secreted substance.

[0004] International Publication No. 2022 / 190733

[0005] Nature Communications, 15 Jun 2023, 14(1): 3567

[0006] In the prior art, there are cases where the influence of the total amount of secretion is not considered, and there is a problem that the characteristics of the secretion cannot be accurately evaluated. Therefore, there has been a situation where further development of technology for accurately evaluating the characteristics of secretions is required.

[0007] In view of such a situation, the main object of this technology is to provide a technology that can accurately evaluate the characteristics of secretions.

[0008] In response to this, the inventors of the present invention have found that the above problem can be solved by a specific secretion evaluation method and secretion evaluation system. Specifically, the present technology provides a secretion evaluation method comprising: a position information identification step of identifying the position information of a secretion cell and an antigen-holding particle based on optical information relating to at least one secretion cell and a carrier holding at least one antigen-holding particle; a secretion information acquisition step of acquiring information about a secretion that binds to an unbinding molecule held by the secretion cell and the antigen-holding particle; and a secretion evaluation step of evaluating the characteristics of the secretion based on the position information and the information about the secretion.

[0009] Furthermore, this technology also provides a particle analyzer comprising: a flow channel through which a liquid containing particles flows; an irradiation unit that irradiates light onto particles flowing through the flow channel; and a detection unit that detects the light generated by the light irradiation; and a processing device that performs: a position information identification step of identifying the position information of a secretory cell and an antigen-containing particle based on optical information relating to a carrier holding at least one secretory cell and at least one antigen-containing particle; a secretion information acquisition step of acquiring information on secretions that bind to unbound molecules possessed by the secretory cell and the antigen-containing particle; and a secretion evaluation step of evaluating the characteristics of the secretions based on the position information and the information on the secretions.

[0010] This figure shows an example of the flow of the secretion evaluation method according to the first embodiment. This figure shows an overview of the judgment algorithm in case a. This figure shows an overview of the judgment algorithm in case b. This figure shows the relationship between the amount of antigen-antibody complex [AB] when the amount of antibody [B] is given. This figure shows an overview of the judgment algorithm in case c. This figure shows a schematic of the case where non-antigen-containing particles are present in addition to secretory cells and antigen-containing particles. This figure shows an overview of the judgment algorithm in the case of a functional assay. This figure shows an overview of a judgment algorithm different from that in Figure 7 in the case of a functional assay. This figure shows the relationship between the fluorescence detection amount and the time change of the flowing secretory cells, antigen-containing particles, and secretions in phenomena to be detected and phenomena not to be detected. This figure shows how the overlap rate between antigen-containing particles and secretions is quantified. This figure shows how the overlap rate between secretory cells and secretions is quantified. This figure shows how the overlap rate is quantified using two-dimensional image data. This figure shows how the overlap between antigen-containing particles and secretions and the overlap between secretory cells and secretions are mapped in two dimensions. This figure shows a schematic of the case where differences in fluorescence intensity are taken into consideration. This diagram illustrates the diagram shown in Figure 14, projected into one dimension. This diagram explains the method for calculating the overlap of each fluorescence signal. This diagram schematically shows the overall configuration of the secretion evaluation system 1000 according to this embodiment. This diagram shows specific example 1 of gate display. This diagram shows specific example 2 of gate display. This diagram shows specific example 3 of gate display. This diagram shows specific example 4 of gate display. This diagram shows specific example 5 of gate display. This diagram shows the assumptions in a specific example of true positive rate presentation. This diagram shows an overview of an example of true positive rate determination. This diagram shows a specific example of true positive rate presentation. A is a diagram showing the reaction between cells and secretions, and B is a diagram showing the pattern of the signal detection method. This diagram shows an overview of the determination algorithm in case d. This diagram explains how the user selects a gate. This diagram explains an overview of automatic gate selection. This diagram explains the area portion of each sample that is both Receiver+ and Secretion+. This flowchart is an example of an embodiment of a method that utilizes a single feature.This diagram illustrates a method for finding the conditions under which the "number of features matching" increases sharply. This flowchart shows an example of a specific embodiment of the method for finding points of sharp increase. A is a diagram illustrating the area portion of each sample that is both Receiver+ and Secretion+, and B is a diagram illustrating each aspect of the area that is Secretion+ for each sample. This flowchart shows an example of an embodiment of a method that utilizes two features. This diagram illustrates an example of the process up to gate determination. This diagram illustrates an example of the process up to gate determination.

[0011] The following describes preferred embodiments for implementing this technology. Note that the embodiments described below represent typical embodiments of this technology, and the scope of this technology is not limited to these embodiments. The description of this technology will proceed in the following order: 1. First Embodiment (Secretion Evaluation Method) (1) Description of the First Embodiment (1-1) Description of the Conventional Technology (1-2) Overview of the Technology (2) Example of the First Embodiment (2-1) Position Information Identification Process S101 (2-2) Secretion Information Acquisition Process S102 (2-3) Secretion Evaluation Process S103 (2-4) Photodetection Process S104 (2-5) Captured Molecule Arrangement Process S105 (2-6) Separation Process S106 (2-7) Binding State Detection Process S107 (2-8) Other Processes (3) Specific Examples of Gate Display (3-1) Specific Example 1 (3-2) Specific Example 2 (3-3) Specific Example 3 (3-4) Specific Example 4 (3-5) Specific Example 5 (4) Specific Example of True Positivity Rate Presentation 2. Second Embodiment (Secretion Evaluation System 1000) (1) Description of the Second Embodiment (2) Example of the Second Embodiment (2-1) Particle Analyzer 100 (2-2) Processing Device 200

[0012] 1. First Embodiment (Method for Evaluating Secretions)

[0013] (1) Description of the first embodiment

[0014] The secretion evaluation method according to this embodiment includes a location information identification step S101, a secretion information acquisition step S102, and a secretion evaluation step S103. Additionally, a photodetection step S104, a capture molecule placement step S105, a separation step S106, a binding state detection step S107, etc., may be performed as needed.

[0015] (1-1) Explanation of the prior art

[0016] While there are conventional techniques for evaluating secretions from cells that detect only secretions specific to unbound molecules, these techniques do not take into account the influence of the total amount of secretions, and therefore cannot accurately evaluate properties such as affinity and specificity based on secretions alone.

[0017] For example, in Non-Patent Document 1 mentioned above, the measurement target is the region excluding the cell region from the entire carrier region that holds the cells, and only the amount of secretion showing specific binding is evaluated. In other words, the total amount of secretion is not considered. Therefore, it was not possible to accurately evaluate characteristics such as binding affinity and binding specificity of secretions. Specifically, with a technique that only detects the amount of bound secretion, it was impossible to distinguish between events with low binding ability and high secretion volume and events with high binding ability and low secretion volume.

[0018] Based on the above, when interpreting the amount of binding secretion signal on unbinding molecules, it was impossible to distinguish between cases where the secretion was highly specific and low in volume, and cases where the secretion was low in specificity and high in volume. Therefore, there was a problem in that the characteristics of the secretion could not be evaluated with high accuracy.

[0019] (1-2) Overview of this technology

[0020] This technology includes a position information identification step S101 that identifies the position information of a secretory cell and an antigen-holding particle based on optical information relating to at least one secretory cell and a carrier holding at least one antigen-holding particle; a secretion information acquisition step S102 that acquires information about secretions that bind to unbinding molecules held by the secretory cell and the antigen-holding particle; and a secretion evaluation step S103 that evaluates the characteristics of the secretions based on the position information and the information about the secretions, thereby improving the accuracy of secretion evaluation.

[0021] In other words, as will be described later, this technology enables distinguishable measurement of binding molecules that specifically bind to unbinding molecules and binding molecules that nonspecifically bind to unbinding molecules by attaching positional information within the carrier to both. Furthermore, by setting the staining pattern of the light signal to be measured, or the quantitative value of the light and an evaluation index based on that quantitative value, it becomes possible to evaluate and determine the amount of unbound secretions, specific binding, and binding affinity. Therefore, based on the evaluation index, it becomes possible to discover and sort secretory cells that secrete secretions with high binding affinity and high binding specificity with high accuracy.

[0022] This technology enables highly accurate identification of secretions that specifically bind to unbound molecules during screening, and allows for highly accurate identification and sorting of secretory cells that secrete secretions that specifically bind to unbound molecules. Furthermore, by excluding secretions that exhibit high secretion levels but low specific binding to unbound molecules, the false positive rate can be reduced. This reduces the cost, time, and effort required for processes after screening. In addition, the true positive rate, which consists of secretions that exhibit high specific binding to unbound molecules, can be increased. Therefore, since only the target secretory cells can be obtained, the number of true positives among the total number obtained can be increased.

[0023] Furthermore, by improving the accuracy of evaluating the binding affinity and binding specificity of secretions, it becomes possible to obtain secretions with high binding affinity and binding specificity with high accuracy.

[0024] In this specification, "secretion" refers to a general term for substances secreted from secretory cells, such as antibodies and cytokines. "Binding molecule" refers to a general term for molecules that exhibit binding ability, including antibodies, such as antibodies, immunoglobulins, antibody fragments, Fab, scFv, VHH, binding peptides, affibody, aptamers, and binding nucleic acids. Furthermore, "unbound molecule" refers to a general term for molecules that are bound, such as antigens and receptors. In addition, "binding molecule that exhibits specific binding to unbound molecules" refers to, for example, antigen-specific antibodies and receptor-specific binding cytokines.

[0025] (2) An example of the first embodiment

[0026] Figure 1 shows an example of a flow chart of the secretion evaluation method according to the first embodiment. As shown in Figure 1, the secretion evaluation method according to this embodiment includes at least: a position information identification step S101 that identifies the position information of the secreting cell and the antigen-holding particle based on optical information relating to at least one secreting cell and a carrier holding at least one antigen-holding particle; a secretion information acquisition step S102 that acquires information about the secretion that binds to the unbinding molecule held by the secreting cell and the antigen-holding particle; and a secretion evaluation step S103 that evaluates the characteristics of the secretion based on the position information and the information about the secretion. Each step will be described below.

[0027] (2-1) Location information identification step S101

[0028] In the position information identification step S101, the position information of the secretory cell and the antigen-containing particle is identified based on optical information relating to at least one secretory cell and at least one carrier holding an antigen-containing particle. An example of the position information identification step S101 is described below.

[0029] In this specification, "secretory cell" refers to a cell that produces and secretes a specific substance (i.e., a secretion) inside or outside the body. Specifically, examples include immune system cells, endocrine cells, exocrine cells, neurosecretory cells, enterocrine cells, etc., but in this embodiment, it is particularly immune system cells. Examples of immune system cells include innate immune system cells such as macrophages, neutrophils, dendritic cells, eosinophils, basophils, and natural killer cells; and adaptive immune system cells such as T cells (e.g., helper T cells, killer T cells, regulatory T cells, etc.), B cells, plasma cells, and memory cells, but this embodiment is not limited to these.

[0030] In this specification, "antigen-bearing particles" refer to particles that play a role in bearing antigens on their surface. Specifically, examples include antigen-presenting cells such as dendritic cells, macrophages, and B cells; cells capable of presenting antigens under specific conditions, such as virus-infected cells, tumor cells, epithelial cells, endothelial cells, fibroblasts, vascular endothelial cells, cells expressing antigens by genetic recombination, and antigen gene knock-in cells; and antigen carriers, which are antigens attached to the surface of carriers capable of bearing antigens (e.g., proteins, polymers, nanoparticles, liposomes, etc.), but are not limited to these in this embodiment. Furthermore, in this specification, "antigen-non-bearing particles" refer to particles that do not possess antigens. Specifically, examples include antigen-non-bearing cells, but are not limited to these in this embodiment. Antigen-non-bearing cells can be produced by knocking out or knocking down the antigen gene in the cell. In addition, in this embodiment, the shape of the particles may be spherical, approximately spherical, approximately elliptical, non-spherical, etc., but are not limited to these in this embodiment.

[0031] In this specification, "carrier" means a device that holds cells, and may hold one or more cells. For example, it may hold biological components (e.g., cell-derived components (e.g., secretions)). Holding biological components in the carrier includes, for example, cases where the biological components are captured by the carrier or encapsulated by the carrier. The carrier may be, for example, a carrier used for secretion analysis. The carrier may also be an emulsion. In this case, cells or particles may be separated while contained within the emulsion. Furthermore, in this case, the separated product may be an emulsion, and the dispersed phase constituting the emulsion may be particles contained within the emulsion containing the particles to be separated. The dispersion medium constituting the emulsion may be appropriately selected, for example, depending on the type of emulsion. In addition, the emulsion may be a multiple emulsion. Examples of multiple emulsions include oil-in-water ("o / w / o") and water-in-oil ("w / o / w") emulsions. The shape of the carrier is not particularly limited and may be any shape such as a roughly spherical shape, a roughly elliptical shape, or a rod shape. In addition to the emulsion described above, the carrier can also be beads, gel capsules, etc., but is not limited to these in this embodiment.

[0032] Optical information relating to the carrier is acquired in the photodetection step S104, which will be described later. Examples of optical information relating to the carrier include images or fluorescence information. Therefore, in this embodiment, the positional information of the secretory cells and the antigen-containing particles is identified based on these images or fluorescence information.

[0033] Examples of images include fluorescence images, bright-field images, dark-field images, phase-contrast images, fluorescence resonance energy images, illumination interference images, transmission electron microscope images, scanning electron microscope images, and multispectral images, but in this embodiment, fluorescence images are particularly important.

[0034] Fluorescence information can include, for example, fluorescence labeling information and fluorescence signal information. Specifically, it can include, for example, light intensity data of light containing fluorescence (which may also be feature quantities such as Area, Height, and Width).

[0035] A method for identifying location information based on a fluorescent image or fluorescence information includes, for example, labeling the secretory cells and antigen-containing particles with an antibody labeled with a fluorescent dye (e.g., FITC (fluorescein isothiocyanate), PE (perlidenine erythrocyanine), APC (allophyllometric blue), etc.), a fluorescent staining reagent (e.g., Calcein, Calcein AM, Calcein Blue, FDA (fluorescein diacetate), etc.), and then identifying the location information based on the fluorescence image or fluorescence information possessed by the secretory cells and antigen-containing particles. This makes it possible to determine the location of the secretory cells and antigen-containing particles within the carrier.

[0036] (2-2) Secretion information acquisition step S102

[0037] In the secretion information acquisition step S102, information is acquired regarding secretions that bind to unbound molecules held by the secretory cells and the antigen-bearing particles. An example of the secretion information acquisition step S102 is described below.

[0038] In this specification, "secretion" includes all substances secreted from cells. Specifically, this includes antibodies, antibody fragments, aptamers (e.g., nucleic acid aptamers, peptide aptamers, etc.), molecularly imprinted polymers, exosomes, extracellular vesicles, neurotransmitters, enzymes, signaling molecules (e.g., cytokines, chemokines, etc.), hormones, digestive fluids, etc. However, in this technology, it is particularly one or more selected from the group consisting of antibodies, antibody fragments, aptamers, and molecularly imprinted polymers, and more particularly antibodies.

[0039] In the secretion information acquisition step S102, for example, after acquiring information on whether the secretion remains on the secretory cell, when the secretion is not detected on the secretory cell, it is determined whether the secretory cell is (-), or when the secretion is detected on the secretory cell, it is determined whether the secretory cell is (+), and when the secretion is not detected on the antigen-carrying particle, it is determined whether the antigen-carrying particle is (-), or when the secretion is detected on the antigen-carrying particle, information (i) on whether the antigen-carrying particle is (+) is acquired.

[0040] In this embodiment, for the secretion as well, as described in the position information specifying step S101, it is labeled using an antibody labeled with a fluorescent dye, a fluorescent staining reagent, or the like. Thereby, it is possible to grasp at which position in the carrier the secretion exists.

[0041] Next, in the secretion information acquisition step S102, the difference in signals between the unstained secretion and the stained secretion, or the signal ratio of the unstained secretion and the stained secretion, on the secretory cell and the antigen-carrying particle is acquired as information (ii) regarding the secretion that binds to the antigen presented by the secretory cell and the antigen-carrying particle.

[0042] Also, in this case, in the secretion information acquisition step S102, an evaluation index consisting of subtraction or division of the unstained secretion can be further used.

[0043] Then, based on the information (i) and information (ii) acquired in the secretion information acquisition step S102, the secretion is evaluated in the secretion evaluation step S103. Regarding a specific example of the information (i) and information (ii), in the secretion evaluation step S103, it will be described together with the evaluation of the secretion derived from these information.

[0044] (2-3) Secretion evaluation step S103

[0045] In the secretion evaluation step S103, based on the position information and the information regarding the secretion (particularly, the information (i) and information (ii)), the characteristics of the secretion are evaluated. An example of the secretion evaluation step S103 will be described below.

[0046] <Case a shown in FIG. 2 - Information (i)>

[0047] Figure 2 shows an outline of the determination algorithm in the case of case a. In the secretion information acquisition step S102 described above, in case a where the secretion does not remain on the secretory cell, for the following three cases: - When the secretory cell is (-) and the antigen-carrying particle is (+). - When the secretory cell is (+) and the antigen-carrying particle is (+). - When the secretory cell is (-) and the antigen-carrying particle is (-). Information (i) regarding the secretion that binds to the binding molecules carried by the secretory cell and the antigen-carrying particle can be obtained.

[0048] And for these three cases: - When the secretory cell is (-) and the antigen-carrying particle is (+), the specificity is high and it is "positive". - When the secretory cell is (+) and the antigen-carrying particle is (+), the specificity is low and it is "negative". - When the secretory cell is (-) and the antigen-carrying particle is (-), there is no secretion, or although secretion occurs, it does not bind specifically to the secretory cell and the antigen-carrying particle, so it is "negative". The evaluation of the secretion can be performed respectively.

[0049] <Case a shown in Figure 2 - Information (ii)>

[0050] In the secretion information acquisition step S102 described above, as a specific example of the information (ii), after creating a sample of non-stained secretion, the difference in signals (particularly fluorescence signals) between the non-stained secretion and the stained secretion (for example, stained secretion - non-stained secretion), or the signal ratio (particularly fluorescence signal) of the non-stained secretion and the stained secretion (for example, stained secretion / non-stained secretion) is obtained.

[0051] In this case, the following can be evaluated as "positive": • Stained secretion - Unstained secretion (in secretory cells) < Stained secretion - Unstained secretion (in antigen-containing particles), • Stained secretion / Unstained secretion (in secretory cells) < Stained secretion / Unstained secretion (in antigen-containing particles), or • Stained secretion - Unstained secretion (in antigen-containing particles) / Stained secretion - Unstained secretion (in secretory cells) > Any value. In this embodiment, the numerical value used for evaluation may be, for example, the average value within a cell region, the density of a cell region, the sum of signals in a cell region, the maximum signal value in a cell region, or the median signal value in a cell region.

[0052] Based on the above, in case a (where secretions do not remain on the surface of secretory cells), it is possible to identify secretions that show specific binding to unbound molecules based on the secretion binding pattern. Furthermore, by using the evaluation described above as an indicator, it is possible to identify secretions that show specific binding to unbound molecules based on this indicator.

[0053] Here, in the case of low specificity, if the binding of secretions that show low specificity to unbound molecules to both secretory cells and antigen-carrying particles is similar, the stained / unstained secretions will increase to a similar extent for both secretory cells and antigen-carrying particles. Therefore, it is possible that cases of antigen-carrying particles (+) may include cases of secretory cells (+) and antigen-carrying particles (+), and may be evaluated as "positive" due to low specificity. For this reason, as described in the secretion information acquisition step S102, by further using an evaluation index consisting of subtraction or division of unstained secretions to exclude cases of secretory cells (+) and antigen-carrying particles (+), it becomes possible to determine cases of secretory cells (-) and antigen-carrying particles (+), which show high specificity, as the highest priority "positive". In this case, even if the secretory cells (-) and antigen-carrying particles (+) are not entirely secretory cells (-), those with a tendency towards low secretory cell values ​​and high antigen-carrying particle values ​​can be given priority.

[0054] Furthermore, in the case of low specificity, if the binding of secretions to secretory cells and antigen-containing particles differs between the two, and the secretions exhibit low specific binding to unbinding molecules, then in the position information identification step S101, the carrier may further hold at least one antigen-free particle, and the position information of the antigen-free particle may be further identified based on the optical information relating to the carrier, and the position information of the antigen-free particle may be used in the secretion information acquisition step S102.

[0055] In other words, a third type of cell, the antigen-free particle, is added and used as a negative control. In this case, it is preferable that the antigen-possessing particle and the antigen-free particle are identical in all conditions except for whether or not they present the antigen. For example, if the antigen-possessing particle is a cell, it is preferable that the antigen-free particle is from the same cell line as the antigen-possessing particle.

[0056] When a third type of cell, antigen-free particles, is added, in the secretion information acquisition step S102, the difference in signals (particularly fluorescence signals) between the antigen-free particles and the antigen-possessing particles (e.g., antigen-possessing particles - antigen-free particles), or the signal (particularly fluorescence signal) ratio of the antigen-free particles and the antigen-possessing particles (e.g., antigen-possessing particles / antigen-free particles), is acquired as information (ii) regarding secretions bound to unbound particles possessed by the secretory cells and the antigen-possessing particles.

[0057] In this case, the following can be evaluated as "positive": • Antigen-containing particle - Antigen-non-containing particle (in secretory cells) < Antigen-containing particle - Antigen-non-containing particle (in antigen-containing particle), • Antigen-containing particle / Antigen-non-containing particle (in secretory cells) < Antigen-containing particle / Antigen-non-containing particle (in antigen-containing particle), or • Antigen-containing particle - Antigen-non-containing particle (in antigen-containing particle) / Antigen-containing particle - Antigen-non-containing particle (in secretory cells) > any value.

[0058] Based on the above, by more accurately measuring nonspecific binding, specific binding can be evaluated more accurately.

[0059] <In the case of case b shown in Figure 3 - Information (i)>

[0060] Figure 3 shows an overview of the determination algorithm for case b. In case b, where the secretion remains on the secretory cell, there are no cases where the secretory cell (-) and antigen-containing particle (+) are present. Furthermore, in the case of the secretory cell (+) and antigen-containing particle (+), it can result in a highly specific "positive" or a low specific "negative" or "positive," so it cannot be determined in the same way as in case a described above. Therefore, in case b, the determination will be made using the information (ii) described below.

[0061] <In the case of case b shown in Figure 3 - Information (ii)>

[0062] In the secretion information acquisition step S102 described above, as a specific example of the information (ii), a sample of unstained secretions is prepared, and the difference in signals (particularly fluorescence signals) between the unstained secretions and the stained secretions (e.g., stained secretions - unstained secretions), or the signal (particularly fluorescence signal) ratio of the unstained secretions and the stained secretions (e.g., stained secretions / unstained secretions) is acquired.

[0063] In this case, the following can be evaluated as "positive": • Stained secretion - Unstained secretion (in secretory cells) < Stained secretion - Unstained secretion (in antigen-containing particles), • Stained secretion / Unstained secretion (in secretory cells) < Stained secretion / Unstained secretion (in antigen-containing particles), or • Stained secretion - Unstained secretion (in antigen-containing particles) / Stained secretion - Unstained secretion (in secretory cells) > Any value. In this embodiment, the numerical value used for evaluation may be, for example, the average value within a cell region, the density of a cell region, the sum of signals in a cell region, the maximum signal value in a cell region, or the median signal value in a cell region.

[0064] Based on the above, in case b (where secretions remain on the surface of secretory cells), the evaluation described above can be used as an indicator to identify secretions that show specific binding to unbound molecules.

[0065] Here, in the case of low specificity, if the binding of secretions that show low specificity to unbound molecules to both secretory cells and antigen-carrying particles is similar, then, as in case a described above, the stained / unstained secretions will increase to a similar extent for both secretory cells and antigen-carrying particles. Therefore, it is possible that cases of antigen-carrying particles (+) may include cases of secretory cells (+) and antigen-carrying particles (+), and these may be evaluated as "positive" due to low specificity. Therefore, as described in the secretion information acquisition step S102, by further using an evaluation index consisting of subtraction or division of unstained secretions to exclude cases of secretory cells (+) and antigen-carrying particles (+), it becomes possible to prioritize the determination of cases of secretory cells (-) and antigen-carrying particles (+) that show high specificity as "positive". In this case, even if the cases of secretory cells (-) and antigen-carrying particles (+) are not entirely secretory cells (-), those with a tendency towards low secretory cell values ​​and high antigen-carrying particle values ​​can be given priority.

[0066] Furthermore, in the case of low specificity, if the binding of secretions to secretory cells and antigen-containing particles differs between the two, and the secretions exhibit low specific binding to unbound molecules, then, similar to case a described above, in the position information identification step S101, the carrier further holds at least one antigen-free particle, and the position information of the antigen-free particle is further identified based on the optical information relating to the carrier, and the position information of the antigen-free particle can be used in the secretion information acquisition step S102.

[0067] In other words, a third type of cell, the antigen-free particle, is added and used as a negative control. In this case, it is preferable that the antigen-containing particle and the antigen-free particle are identical in all conditions except for whether or not they contain an antigen. For example, if the antigen-containing particle is a cell, it is preferable that the antigen-free particle is from the same cell line as the antigen-containing particle.

[0068] When a third type of cell, antigen-free particles, is added, in the secretion information acquisition step S102, the difference in signals (particularly fluorescence signals) between the antigen-free particles and the antigen-possessing particles (e.g., antigen-possessing particles - antigen-free particles), or the signal (particularly fluorescence signal) ratio of the antigen-free particles and the antigen-possessing particles (e.g., antigen-possessing particles / antigen-free particles), is acquired as information (ii) regarding secretions bound to unbound particles possessed by the secretory cells and the antigen-possessing particles.

[0069] In this case, the following can be evaluated as "positive": • Antigen-containing particle - Antigen-non-containing particle (in secretory cells) < Antigen-containing particle - Antigen-non-containing particle (in antigen-containing particle), • Antigen-containing particle / Antigen-non-containing particle (in secretory cells) < Antigen-containing particle / Antigen-non-containing particle (in antigen-containing particle), or • Antigen-containing particle - Antigen-non-containing particle (in antigen-containing particle) / Antigen-containing particle - Antigen-non-containing particle (in secretory cells) > any value.

[0070] Based on the above, by more accurately measuring nonspecific binding, specific binding can be evaluated more accurately.

[0071] <In the case of the functional assay shown in Figure 7>

[0072] Figure 7 shows an overview of the determination algorithm for a functional assay. In the functional assay shown in Figure 7, the signal detection is not limited to detecting the amount of binding of secretions secreted by secretory cells to the target antigen on the antigen-carrying particles, but also includes signals that change when secretions (e.g., antibodies, ligands, etc.) bind to the target antigen expressed on the antigen-carrying particles. That is, in the secretion evaluation step S103, the signal that changes when the secretions bind to the unbinding molecules on the antigen-carrying particles is detected, and the detected signal can be used for the evaluation.

[0073] Specifically, methods that use a fluorescent signal that changes depending on the amount of secretion (in this case, antibody) bound to an antigen include, for example, the direct antibody method, the indirect antibody method, the enzymatic method, and the FRET method. Specifically, these methods include methods that use a fluorescently labeled antibody (not limited to antibodies, but may be any binding molecule) that binds to an antibody to detect the fluorescent signal, and methods that use an enzyme-labeled antibody that binds to an antibody to amplify and accumulate the fluorescent signal with an enzyme and then detect the fluorescent signal.

[0074] Other methods include, for example, inhibition methods and competition methods. These methods detect the fluorescence signal produced by the antibody by adding another binding molecule that binds to the same antigen before, simultaneously with, or after the antibody has bound to the antigen.

[0075] In the functional assay shown in Figure 7, the following methods can be used to detect the changing signal: • A method of detecting the fluorescent signal using a fluorescently labeled antibody or enzyme-labeled antibody similar to the direct antibody method, indirect antibody method, enzymatic method, FRET method, etc., after performing an inhibition method or a competitive method; or • A method of detecting the fluorescent signal by using another binding molecule that binds to the antigen and has been fluorescently labeled. These methods can be used as indicators for the evaluation. The indicators for the evaluation can be calculated in the same way as in case a described above. Note that either the fluorescent signal detected in case a or the signal detected in the functional assay shown in Figure 7 may be used as the indicators for the evaluation, or both the fluorescent signal detected in case a and the signal detected in the functional assay shown in Figure 7 may be used as the indicators for the evaluation. Furthermore, the indicators for the evaluation of the signal detected in the functional assay shown in Figure 7 can be calculated in the same way as the indicators for the evaluation described in case a.

[0076] <In the case of the functional assay shown in Figure 8>

[0077] Figure 8 shows an overview of a different determination algorithm for a functional assay compared to Figure 7. In the functional assay shown in Figure 8, as with the functional assay shown in Figure 7, the signal detection is not limited to detecting the amount of binding of secretions secreted by secretory cells to the target antigen held on the antigen-carrying particles, but also includes signals that change when secretions (e.g., antibodies, ligands, etc.) bind to the target antigen expressed on the antigen-carrying particles. That is, as with the functional assay shown in Figure 7, in the secretion evaluation step S103, the signal that changes when the secretions bind to the unbinding molecules held on the antigen-carrying particles is detected, and the detected signal can be used for the evaluation.

[0078] Specifically, this method detects function-dependent signal changes that occur in response to the binding of secretions (in this case, antibodies) to an antigen. Examples include calcium assays, reporter assays, internalization assays, and cAMP assays.

[0079] In the functional assay shown in Figure 8, the assay described above can be used as the method for detecting the changing signal, and this can be used as the evaluation index. The evaluation index can be calculated in the same way as in case a described above. Either the fluorescent signal detected in case a or the signal detected in the functional assay shown in Figure 8 may be used as the evaluation index, or both the fluorescent signal detected in case a and the signal detected in the functional assay shown in Figure 8 may be used as the evaluation index. Furthermore, the evaluation index for the signal detected in the functional assay shown in Figure 8 can be calculated in the same way as the evaluation index described in case a.

[0080] (2-4) Photodetection process S104

[0081] The secretion evaluation method according to this embodiment may further include a photodetection step S104. In the photodetection step S104, light from the carrier is detected and optical information relating to the carrier is acquired. An example of the photodetection step S104 is described below.

[0082] The optical information relating to the carrier may include, for example, an image or fluorescence information, as described in the position information identification step S101. Since the image or fluorescence information is the same as described in "position information identification step S101," a detailed explanation is omitted here.

[0083] In the photodetection step S104, the irradiation unit 101 irradiates the carriers flowing through the channel C (particularly the detection region) with light (for example, excitation light), and the detection unit 102 detects the light generated by the light irradiation.

[0084] The irradiation unit 101 irradiates the carriers flowing in the channel C with light (e.g., excitation light). The irradiation unit 101 includes a light source unit that emits light and a light guide optical system that guides the light to the irradiation point. The light source unit includes one or more light sources. Examples of light sources include laser light sources and LEDs. The wavelength of the light emitted from each light source may be ultraviolet light, visible light, or infrared light. The light guide optical system includes optical components such as beam splitter groups, mirror groups, and optical fibers. The light guide optical system may also include lens groups for focusing light, such as objective lenses. There may be one or more irradiation points where the sample S and light intersect. The irradiation unit 101 may also be configured to focus light irradiated from one or more different light sources to a single irradiation point.

[0085] The detection unit 102 includes at least one photodetector that detects light generated by light irradiation of a carrier. The light to be detected is, for example, fluorescence or scattered light (for example, one or more of forward scattered light, back scattered light, and side scattered light). Each photodetector includes one or more light-receiving elements, for example, a light-receiving element array. Each photodetector may include one or more PMTs (photomultiplier tubes) and / or photodiodes such as APDs and MPPCs as light-receiving elements. The photodetector may include, for example, a PMT array in which multiple PMTs are arranged in a one-dimensional direction. The detection unit 102 may also include an image sensor such as a CCD or CMOS. The detection unit 102 can acquire an image (for example, a bright-field image, a dark-field image, a fluorescence image, etc.) as optical information regarding the carrier using the image sensor.

[0086] The detection unit 102 includes a detection optical system that directs light of a predetermined detection wavelength to a corresponding photodetector. The detection optical system includes a spectral section such as a prism or diffraction grating; and a wavelength separation section such as a dichroic mirror or optical filter. The detection optical system is configured, for example, to spectrally separate the light generated by irradiating a carrier with light, and to detect the spectrally separated light using a plurality of photodetectors, more than the number of fluorescent dyes to which the carriers are labeled. The detection optical system is also configured, for example, to separate the light corresponding to the fluorescence wavelength range of a specific fluorescent dye from the light generated by irradiating a carrier with light, and to have the separated light detected by the corresponding photodetector.

[0087] Furthermore, the detection unit 102 may include a signal processing unit that converts the electrical signal obtained by the photodetector into a digital signal. The signal processing unit may include an A / D converter as the device that performs the conversion. The digital signal obtained by the conversion by the signal processing unit may be transmitted to the information processing unit 103. The digital signal may be treated by the information processing unit 103 as optical information relating to the carrier (hereinafter also referred to as "optical data"). The optical data may be, for example, optical data including fluorescence data. Specifically, the optical data may be light intensity data, and the light intensity may be light intensity data of light including fluorescence (it may also be feature quantities such as Area, Height, Width).

[0088] Furthermore, it is generally known that in light detection by the detection unit 102, a shift in focus relative to the object to be detected can affect the fluorescence intensity, potentially causing fluctuations in fluorescence brightness. In contrast, the light detection step S104 detects fluorescence from the fluorescently labeled carrier, and in the position information identification step S101 described above, the optical information related to the carrier can be corrected based on the detected fluorescence intensity of the carrier.

[0089] Specifically, the carrier is fluorescently labeled, and the peak or integrated value of the detected fluorescence signal is referenced. Using the fluorescence signal peak provides higher sensitivity and sensitivity to focus, while using the integrated value provides more robust characteristics. In this embodiment, it is also possible to combine the fluorescence signal peak and integrated value.

[0090] Furthermore, if the fluorescence intensity due to the fluorescent labeling of the carrier is kept constant, and the fluorescence detection conditions (e.g., excitation light power, excitation light exposure time, etc.) are the same, then variations in fluorescence intensity can be attributed to focus shifts, and a fluorescence correction coefficient due to focus shifts can be calculated.

[0091] Assuming that a similar degree of focus shift occurs in the cells or particles within the carrier, the variation in fluorescence intensity of the cells or particles due to the focus shift can be corrected by applying a fluorescence correction coefficient due to the focus shift to the fluorescence intensity of the cells or particles. Similarly, the fluorescence intensity of secretions and the carrier surface can also be corrected. Therefore, in this embodiment, in the position information identification step S101 described above, a fluorescence correction coefficient can be calculated from the fluorescence intensity of the carrier detected in the photodetection step S104, and one or more fluorescence intensities selected from the group consisting of secretory cells, antigen-containing particles, secretions, and the carrier surface can be corrected.

[0092] Furthermore, in addition to correcting the fluorescence intensity of the fluorescently labeled carrier, the carrier's shape (e.g., length, aspect ratio) can also be used to correct for focus errors. Correction of the fluorescence intensity of cells or particles can be achieved by applying the correction to the entire fluorescence image, or by applying the correction to evaluation metrics used during analysis (e.g., the sum of fluorescence signals, the sum of fluorescence signals within the mask, etc.).

[0093] Furthermore, when correcting focus shifts with unlabeled carriers, methods such as referring to the scattering intensity by the carriers or referring to the degree of blurring in the carrier morphology detection information (e.g., a decrease in the spatial frequency of the contour) can be used. Then, the amount of focus shift can be estimated from the acquired information, and the fluorescence intensity of the detected object can be corrected.

[0094] (2-5) Capture molecule placement step S105

[0095] The secretion evaluation method according to this embodiment may further include a capture molecule placement step S105. In the capture molecule placement step S105, capture molecules that capture the secretion are placed on the carrier. An example of the capture molecule placement step S105 will be described below.

[0096] In the capture molecule arrangement step S105, the amount of secretion that is not bound to the bound molecule is detected. Here, although the secretion secreted from one secretory cell is one type of secretion, in the case of a secretion with high affinity, the amount of bound molecule-bound secretion > the amount of unbound molecule-unbound secretion, and in the case of a secretion with low affinity, the chemical equilibrium is such that the amount of bound molecule-bound secretion << the amount of unbound molecule-unbound secretion. Therefore, in the capture molecule arrangement step S105, the amount of secretion that is not bound to this bound molecule (amount of unbound molecule-unbound secretion) is detected. In this specification, "bound molecule-bound secretion" refers to secretion that is bound to the bound molecule, and "unbound molecule-unbound secretion" refers to secretion that is not bound to the bound molecule.

[0097] Here, assuming that the volume within the carrier is constant, [A], [B], and [AB] are considered to be the amount of unbound molecules (especially antigens) within the carrier, the amount of secretions (binding molecules; especially antibodies) within the carrier, and the amount of complexes (especially antigen-antibody complexes) formed by the unbound molecules and secretions within the carrier, respectively. Even if the dissociation constant Kd = [A][B] / [AB], which indicates the binding affinity of the antigen-antibody reaction, is assumed to be low, if the amount of secretions [B] is low, the amount of complexes [AB] will also be low. Conversely, even if Kd is high, if the amount of secretions [B] is high, the amount of complexes [AB] will also be high. Therefore, there was a problem in that it was not possible to evaluate the affinity of the antigen-antibody reaction when only the amount of complexes [AB] was measured.

[0098] Furthermore, when the bound molecular weight [A] is high (cellular components), the complex amount [AB] is high even if Kd is high (i.e., it corresponds to nonspecific binding). Figure 4 shows the relationship between the amount of secretion and the amount of antigen-antibody complex [AB] when the amount of antibody [B] is high. When the specificity of the antibody amount [B] is low and there are many types of binding partners such as antigen amounts [A'], [A''], [A''''], the signal of the complex amount [AB] bound to the antigen becomes the sum of the complex amounts for each antigen [AB] + [A'B] + [A''B] + [A''''B], so the signal originating from nonspecific binding becomes high.

[0099] In response to this, the above-mentioned problems can be solved by adding a method to detect the amount of unbound secretions (particularly the amount of antigen-unbound antibodies) by placing capture molecules on a carrier to capture secretions. That is, in the capture molecule placement step S105, the amount of unbound secretions is detected. In this embodiment, capture molecules for capturing secretions may be placed on a carrier, and the secretions present on the carrier may be measured.

[0100] In this specification, "capture molecule" includes any molecule that captures secretions. Specifically, it may be, for example, an anti-secretion, and if the secretion is an antibody, it may be a substance that specifically binds to the antibody. Examples of such substances include, but are not limited to, anti-IgG antibodies (including fluorescently labeled anti-IgG antibodies, enzyme-labeled anti-IgG antibodies, and biotin-labeled anti-IgG antibodies), anti-IgA antibodies, anti-IgM antibodies, anti-IgE antibodies, anti-IgD antibodies, anti-light chain antibodies (including anti-κ chain antibodies and anti-λ chain antibodies), antibody ligands, Fab, Fab'2 fragments formed by linking two Fabs, rheumatoid factor, protein A, protein G, protein L, etc.

[0101] <Case c shown in Figure 5>

[0102] Figure 5 shows an overview of the determination algorithm for case c. Specifically, case c is a case in which the carrier is further labeled using an antibody labeled with a fluorescent dye, a fluorescent staining reagent, etc., as in case a. In Figure 5, the labeled carrier is depicted in gray.

[0103] First, the amount of secretion that specifically binds to the soluble molecule and the amount of secretion that is not bound to the soluble molecule are measured. In the capture molecule placement step S105, antigen-containing particles are pre-coexisted within the carrier, and then the amount of secretion that specifically binds to the soluble molecule is measured. Specifically, in order to measure the amount of secretion that specifically binds to the soluble molecule, antigen-containing particles are pre-coexisted within the carrier (corresponding to the above-mentioned soluble molecular weight (particularly antigen amount) [A]), so that secretions that specifically bind to the soluble molecule can be captured on the antigen-containing particles from the secretory cells. Furthermore, by applying a capture molecule (for example, a fluorescently labeled anti-IgG antibody), the amount of secretion that specifically binds to the soluble molecule can be measured (corresponding to the above-mentioned complex amount [AB]).

[0104] Furthermore, in the capture molecule placement step S105, the amount of unbound secretions is measured after the capture molecules have been immobilized in the carrier in advance. Specifically, when secretory cells and antigen-containing particles are encapsulated in a carrier (e.g., PEG, agarose, etc.), in order to measure the amount of unbound secretions, the secretions secreted from the secretory cells can be captured on the carrier by immobilizing a binding molecule (e.g., anti-IgG antibody, etc.) in the carrier in advance, and then the amount of unbound secretions can be measured by applying the capture molecule (e.g., fluorescently labeled anti-IgG antibody, etc.).

[0105] However, for example, since fluorescently labeled anti-IgG antibodies bind to IgG antibodies, they bind to both secretions on carriers and secretions on antigen-bearing particles. Therefore, the total amount of signal (especially the fluorescent signal) within a single event cannot distinguish between secretions that specifically bind to unbound molecules and secretions that bind nonspecifically to unbound molecules using typical flow cytometer data.

[0106] Therefore, by attaching positional information to the signal (particularly the fluorescent signal), it is possible to distinguish between the two (i.e., secretions on the carrier and secretions on the antigen-containing particles). To explain with a specific example, the fluorescently labeled anti-IgG antibody signal located on the antigen-containing particle is treated as the antigen-specific antibody-derived signal, and the fluorescently labeled anti-IgG antibody signal located on the carrier is treated as the antibody-derived signal to distinguish them. When calculating the antibody-derived signal, it may be calculated on the carrier, on the carrier excluding the position of the antigen-containing particle, or on the carrier excluding both the antigen-containing particle and the secreting cell.

[0107] Next, binding affinity and binding specificity are determined by evaluating the difference in signals (especially fluorescent signals) between the amount of secretion that specifically binds to the bound molecule and the amount of secretion that does not bind to the bound molecule, or by the signal (especially fluorescent signal) ratio of the amount of secretion that specifically binds to the bound molecule / the amount of secretion that does not bind to the bound molecule. Specifically, the evaluation can include: - Amount of secretion that specifically binds to the bound molecule - Amount of secretion that does not bind to the bound molecule, or - Amount of secretion that specifically binds to the bound molecule / Amount of secretion that does not bind to the bound molecule. Furthermore, if the indicator exceeds an arbitrarily set threshold, it can be determined that it has high binding affinity or high binding specificity. In addition, as another method of evaluation, relative evaluation can be used, and substances that show a certain percentage or more in the sample can be determined to have high binding affinity or high binding specificity. Note that the numerical value of the secretion amount may be the average value within the cell region, the density of the cell region, the sum of signals in the cell region, the maximum signal value in the cell region, the median signal value in the cell region, etc.

[0108] <Case d shown in Figure 27>

[0109] Figure 27 shows an overview of the determination algorithm for case d. Specifically, case d is a case in which the carrier is further labeled using an antibody labeled with a fluorescent dye, a fluorescent staining reagent, etc., as in case b, and is a method that combines case b and case c. In Figure 27, the labeled carrier is depicted in gray.

[0110] First, the amount of secretion that specifically binds to the soluble molecule and the amount of secretion that is not bound to the soluble molecule are measured. In the capture molecule placement step S105, antigen-containing particles are pre-coexisted within the carrier, and then the amount of secretion that specifically binds to the soluble molecule is measured. Specifically, in order to measure the amount of secretion that specifically binds to the soluble molecule, antigen-containing particles are pre-coexisted within the carrier (corresponding to the above-mentioned soluble molecular weight (particularly antigen amount) [A]), so that secretions that specifically bind to the soluble molecule can be captured on the surface of the antigen-containing particles from the secretory cells. Furthermore, by applying a capture molecule (for example, a fluorescently labeled anti-IgG antibody), the amount of secretion that specifically binds to the soluble molecule can be measured (corresponding to the above-mentioned complex amount [AB]).

[0111] Furthermore, in the capture molecule placement step S105, the amount of unbound secretions is measured after the capture molecules have been immobilized in the carrier in advance. Specifically, when secretory cells and antigen-containing particles are encapsulated in a carrier (e.g., PEG, agarose, etc.), in order to measure the amount of unbound secretions, the secretions secreted from the secretory cells can be captured on the carrier surface by immobilizing capture molecules (e.g., anti-IgG antibodies, etc.) in the carrier in advance. Furthermore, by applying the capture molecules (e.g., fluorescently labeled anti-IgG antibodies, etc.), the amount of unbound secretions can be measured.

[0112] However, for example, since fluorescently labeled anti-IgG antibodies bind to IgG antibodies, they bind to both secretions on the surface of secretory cells and secretions on the surface of antigen-carrying particles. Therefore, the total amount of signal (especially the fluorescent signal) within a single event cannot distinguish between secretions that show specific binding to unbound molecules and secretions that show nonspecific binding to unbound molecules using typical flow cytometer data.

[0113] Therefore, by attaching positional information to the signal (particularly the fluorescent signal), it is possible to distinguish between the two (i.e., secretions from the surface of secretory cells and secretions from the surface of antigen-carrying particles). To explain with a specific example, the fluorescently labeled anti-IgG antibody signal located on the surface of antigen-carrying particles is treated as the antigen-specific antibody-derived signal, and the fluorescently labeled anti-IgG antibody signal located on the surface of secretory cells is treated as the antibody-derived signal to distinguish them. When calculating the antibody-derived signal, the secretory cell surface may be used, the secretory cell surface excluding the position of the antigen-carrying particle, or the secretory cell surface excluding the antigen-carrying particle and the carrier.

[0114] Next, binding affinity and binding specificity are determined by evaluating the difference in signals (especially fluorescent signals) between the amount of secreted material that specifically binds to the unbound molecule and the amount of secreted material from secretory cells, or by the ratio of the signal (especially fluorescent signal) of the amount of secreted material that specifically binds to the unbound molecule / the amount of secreted material from secretory cells. Specifically, the evaluation can include: - Amount of secreted material that specifically binds to the unbound molecule - Amount of secreted material from secretory cells, or - Amount of secreted material that specifically binds to the unbound molecule / Amount of secreted material from secretory cells. Furthermore, if the indicator exceeds an arbitrarily set threshold, it can be determined that it has high binding affinity or high binding specificity. In addition, as another method of evaluation, relative evaluation can be used, and materials that show a value above a certain percentage in the sample can be determined to have high binding affinity or high binding specificity. Note that the numerical value of the secreted material may be the average value within the cell region, the density of the cell region, the sum of signals in the cell region, the maximum signal value in the cell region, or the median signal value in the cell region.

[0115] Based on the above, in case d, by using the evaluation described above as an indicator, it is possible to identify secretions that show specific binding to unbound molecules based on that indicator.

[0116] Therefore, by measuring the amount of unbound molecules in unbound secretions, it becomes possible to evaluate the total amount of secretions, and to calculate evaluation indices based on this evaluation. As a result, it becomes possible to obtain evaluation indices that reflect the binding affinity and binding specificity of secretions with higher accuracy, and to obtain secretions with higher binding affinity and binding specificity with higher accuracy. Furthermore, since the evaluation indices can be calculated based on a single experiment, costs, time, and labor can be reduced.

[0117] <Example of calculating specific binding index and binding affinity index>

[0118] An example of calculating specific binding indices and binding affinity indices is described below. Specifically, a mask for the Carrier region (hereinafter also referred to as the "Cr region") is set based on Bright Field, carrier autofluorescence, or fluorescently stained carrier. Next, masks for the Doner cell region (hereinafter also referred to as the "D region") and the Receiver cell region (hereinafter also referred to as the "R region") are set based on the fluorescence intensity of secretory cells (Doner cell; D) and antigen-holding particles (Receiver cell; R). Then, the Cr-D-R region, obtained by excluding the D and R regions from the Cr region, is defined as the secretion detection region Cr'.

[0119] The sum of the fluorescence intensities in each region is used to define R / Cr' as an index of specific binding of secretions to antigen-containing particles. Alternatively, fluorescence density can be used instead of the sum of fluorescence intensities to define R / Cr' as a specific binding index. A larger R / Cr' index indicates a higher specific binding amount, while a smaller R / Cr' index indicates a lower specific binding amount. Furthermore, the R / Cr' index can also serve as an index of the binding affinity (dissociation index) of secretions to antigen-containing particles. Note that R-Cr' (difference) may be used as both a specific binding index and a binding affinity index instead of R / Cr' (ratio).

[0120] Further specific examples are described below. Figure 6 shows an overview of the case where non-antigen-containing particles coexist with secretory cells and antigen-containing particles. As shown in Figure 6, in addition to secretory cells and antigen-containing particles, non-antigen-containing particles (non-Receiver cells; nonR), which are the same cell type as the antigen-containing particles but do not express the target antigen, can also be coexisted in the Cr region. In this case, R / nonR (ratio) or R-nonR (difference) can be used as the specific binding index and binding affinity index.

[0121] However, if R / nonR or R-nonR is used, information on the amount of secretions from secretory cells is lost. Therefore, by setting the secretion detection region Cr' above a certain threshold, the measurement target can be set according to the amount of secretions. For example, data showing Cr' above a certain threshold can be gated, and R / nonR or R-nonR can be evaluated within that group.

[0122] Furthermore, by fitting it with a calibration curve of the already obtained dissociation constant Kd, the binding affinity index and the dissociation constant Kd can also be calculated based on the binding affinity index. In addition, if the unbound molecules on the antigen-containing particle are detected by fluorescence and denoted as "E", then E can be positioned as an index related to [A], Cr' as [B], and the antigen-containing particle as [AB]. Thus, the dissociation constant index Kd', which corresponds to the dissociation constant (Kd = [A][B] / [AB]), can be calculated based on the following formula (1).

[0123]

[0124] Based on the above, when obtaining target secretions (especially antibodies) through screening, it becomes possible to identify secretions that specifically bind to unbound molecules with high accuracy, and, if necessary, to sort secretory cells that secrete secretions that specifically bind to unbound molecules with high accuracy. Furthermore, by excluding highly secreted, low-specific secretions, the false-positive rate can be reduced, thereby lowering the cost, time, and effort required for processes after screening. In addition, the true positive rate, which is the rate of secretions that show high specific binding to unbound molecules, can be increased, and only the target secretory cells can be obtained, thus increasing the number of true positives out of the total number obtained.

[0125] (2-6) Preparation step S106

[0126] The secretion evaluation method according to this embodiment may further include a sorting step S106. In the sorting step S106, if the characteristics of the secretion are above a certain level, the secretory cells are sorted. An example of the sorting step S106 is described below.

[0127] In this embodiment, the characteristics of the secretion in the secretion evaluation step S103 described above are, in particular, the binding affinity or binding specificity between the binding-tolerant molecule possessed by the antigen-containing particle and the secretion.

[0128] In this specification, "affinity" is an indicator of how strongly a certain secretion (binding molecule) binds to a specific unbinding molecule, and indicates the "strength" of the bond between the secretion and the unbinding molecule. In contrast, "specificity" is an indicator of how selectively a certain secretion binds to a specific unbinding molecule, and indicates the ability of the secretion to select a specific unbinding molecule from among many other candidates, that is, it indicates the "degree of selectivity" in selecting that specific unbinding molecule. When the secretion is an antibody, a particular antibody, due to its high specificity, binds only to a specific antigen, and at the same time, if that particular antibody also has a high affinity for that particular antigen, it will be in a state of being strongly bound to that particular antigen.

[0129] In the sorting step S106, if the binding affinity or binding specificity is above a certain level, the secretory cells are sorted. The evaluation index for binding affinity or binding specificity can be determined, for example, using the specific binding index and binding affinity index calculated in the capture molecule placement step S105 described above. Alternatively, in the binding state detection step S107 described later, the degree to which the secretion is bound to the unbinding molecule possessed by the antigen-bearing particle, which has been quantified, may be used for the determination.

[0130] The method for separating the secretory cells can be a conventionally known method, for example, by using the principle of a cell sorter. Specifically, if the particle analyzer 100 includes a separation unit 104, the information processing unit 103 can determine whether a particle is a particle to be separated (in this case, a secretory cell) based on optical data and / or morphological information. The information processing unit 103 then controls the separation unit 104 based on the result of this determination, and the separation unit 104 can separate the particles to be separated.

[0131] The information processing unit 103 includes, for example, a processing unit that performs processing of various data and a storage unit that stores various data. When the processing unit obtains optical data corresponding to a fluorescent dye from the detection unit 102, it may perform fluorescence leakage correction (compensation processing) on ​​the optical intensity data. In the case of a spectral flow cytometer, the processing unit also performs fluorescence separation processing on the optical data to obtain optical intensity data corresponding to a fluorescent dye. The fluorescence separation processing may be performed, for example, according to the unmixing method described in Japanese Patent Application Publication No. 2011-232259. If the detection unit 102 includes an image sensor, the processing unit may obtain particle morphology information based on the image obtained by the image sensor. The storage unit may be configured to store the acquired optical data. The storage unit may further be configured to store spectral reference data used in the unmixing processing.

[0132] The information processing unit 103 may be configured to output various types of data. For example, the information processing unit 103 may output various types of data (e.g., two-dimensional plots, spectral plots, etc.) generated based on the optical data. The information processing unit 103 may also be configured to accept input of various types of data, for example, to accept gating processing on a plot by a user. The information processing unit 103 may include an output unit (e.g., a display, monitor, personal digital assistant, wearable device, etc.) or an input unit (e.g., a keyboard, voice input device, etc.) for executing such output or input.

[0133] The information processing unit 103 may be configured as a general-purpose computer, for example, as an information processing device equipped with a CPU, RAM, and ROM. The information processing unit 103 may be contained within the housing that houses the irradiation unit 101 and the detection unit 102, or it may be located outside of the housing. Furthermore, various processing or functions performed by the information processing unit 103 may be implemented by a server computer or cloud connected via a network.

[0134] The sorting unit 104 performs sorting of the target particles according to the determination result by the information processing unit 103. The sorting method may be a method in which droplets containing particles are generated by vibration, an electric charge is applied to the droplets of the target particles, and the direction of movement of the droplets is controlled by electrodes, or a method in which the direction of movement of the target particles is controlled within a flow channel structure and sorting is performed. The flow channel structure may be provided with a control mechanism, for example, by pressure (injection or suction) or electric charge. A specific example of the flow channel structure is a chip (for example, the chip described in Japanese Patent Application Publication No. 2020-76736) in which a flow channel C branches downstream into a sorting channel and a waste liquid channel, and specific target particles are sorted into the sorting channel.

[0135] The channel C is configured to allow the sample S to flow through it. The sample S may be a liquid sample containing at least one secretory cell. In particular, the channel C may be configured to form a flow in which at least one secretory cell contained in the sample S is arranged in a substantially straight line. The channel structure including the channel C may be designed to form a laminar flow. In particular, the channel structure may be designed to form a laminar flow in which the flow of the sample S (sample flow) is surrounded by the flow of the sheath liquid. The design of the channel structure may be appropriately selected according to the size of the particles to be separated, and known designs may be adopted. The channel C may be formed in a flow channel structure such as a microchip (a chip having a channel C on the order of micrometers), a flow cell, etc. The width of the channel C is, for example, 1 mm or less, and in particular, 10 μm or more and 1 mm or less. Furthermore, the channel C and the channel structure containing it may be formed from materials such as polycarbonate, cycloolefin polymer, polypropylene, PDMS (polydimethylsiloxane), polymethyl methacrylate (PMMA), polyethylene, polystyrene, glass, and silicon.

[0136] (2-7) Coupling state detection step S107

[0137] The secretion evaluation method according to this embodiment may further include a binding state detection step S107. In the binding state detection step S107, the positional information of the secretion is further identified in the positional information identification step S101, and the state in which the secretion is bound to the molecule to be bound held by the antigen-holding particle is detected based on the relationship between the positional information of the secretory cell, the antigen-holding particle, and the secretion. An example of the binding state detection step S107 will be described below.

[0138] Conventional techniques, such as using a standard flow cytometer, rely on information like Area, Height, and Width to identify particles containing cells. This presents a challenge in determining the positional relationships between particles (including cells) and secretions. While some imaging cytometers can calculate the similarity of target particles using positional information, these methods rely on detecting particle positions as a mask and extracting information from within that mask. Consequently, when particles overlap, only limited positional information can be used, making analysis using the regions of multiple particles impossible.

[0139] In contrast, the above-mentioned problems can be solved by performing the binding state detection step S107. Specifically, in the binding state detection step S107, as described above, in the position information identification step S101, the position information of the secretion is further identified after being labeled with an antibody labeled with a fluorescent dye, a fluorescent staining reagent, etc., and the state in which the secretion is bound to the molecule to be bound held by the antigen-holding particle is detected based on the relationship between the position information of the secretory cell, the antigen-holding particle, and the secretion.

[0140] Specifically, in the binding state detection step S107, the degree to which the secreted substance is bound to the unbound molecule possessed by the antigen-possessing particle is quantified based on the respective fluorescence signals of the secretory cell, the antigen-possessing particle, and the secretion (particularly the antibody). More specifically, the positions of the secretory cell, the antigen-possessing particle, and the secretion of interest (particularly the antibody) are extracted from each fluorescence signal, and the relationship of this positional information makes it possible to determine the phenomenon of interest (i.e., the state in which the secretion is bound to the unbound molecule possessed by the antigen-possessing particle). Furthermore, in the quantification, the overlap of each fluorescence signal is used to make it possible to numerically determine the accuracy of the phenomenon of interest.

[0141] Figure 9 shows the relationship between the fluorescence detection amount and the time change of the flowing secretory cells, antigen-containing particles, and secretions in phenomena to be detected and phenomena to be undetected. Figure 9A shows a graph of the phenomenon to be detected, in which secretions (particularly antibodies) secreted from secretory cells (particularly antibody-producing cells) are bound to antigen-containing particles that possess molecules to which they specifically bind. Figure 9B shows a graph of the phenomenon to be undetected, in which secretions secreted from secretory cells do not want to bind to antigen-containing particles that possess molecules to which they specifically bind. In the binding state detection step S107, these phenomena can be distinguished by quantifying them.

[0142] As a concrete example of quantifying the phenomenon described above, we can cite the method of calculating the overlap rate. The method for calculating the overlap rate will be explained in detail below.

[0143] Figure 10 shows how the overlap rate between antigen-carrying particles and secretions is quantified. As shown in Figure 10, in the case of "positive," since there is secretion in the region of the antigen-carrying particle (Receiver), the overlap rate shown at the bottom of A in Figure 10 is often 1. On the other hand, in the case of "negative," since there is no secretion in part or all of the region of the antigen-carrying particle, the value is often smaller than 1 (0.65 in the example shown in Figure 10). In other words, there is also secretion on the secretory cell (Donor) side, and the above value can be said to fluctuate depending on the positional relationship.

[0144] Figure 10B shows an example of the analysis of simulation data in histogram. The results of this simulation showed that "positive" results were often 1, and "negative" results were often other than 1. Note that these values ​​may fluctuate depending on where the fluorescence signal threshold is set, so in such cases, it is advisable to adjust the threshold accordingly.

[0145] Figure 11 shows how the overlap rate between secretory cells and secretions is quantified as an evaluation index different from that in Figure 10. As shown in Figure 11, in the case of "positive," there is secretion in areas other than the secretory cells (Donor), so the overlap rate shown at the bottom of A in Figure 11 is often a value less than 1. On the other hand, in the case of "negative," there is no secretion in some or all of the areas other than the secretory cells, so it is often 1 (1 in the example shown in Figure 11).

[0146] Figure 11B shows an example of the analysis of simulation data in histogram. The results of this simulation showed that in the case of "positive," the value is often something other than 1, and in the case of "negative," it is often 1. Note that this histogram is displayed as "1.0 - value," so the result shown above is obtained by flipping it horizontally. As with the case shown in Figure 10, these values ​​may fluctuate depending on where the fluorescence signal threshold is set, so in that case, it is advisable to adjust the threshold appropriately.

[0147] Furthermore, while Figures 10 and 11 were quantified based on one-dimensional data (1-channel PMT), as shown in Figure 12, the overlap rate can also be calculated using two-dimensional image data (for example, 32-channel PMT) in the same way as in Figures 10 and 11.

[0148] Furthermore, since the overlap between antigen-carrying particles and secretions, and the overlap between secretory cells and secretions are independent events, they can be mapped in two dimensions, as shown in Figure 13, and represented as a scattergram, similar to those used in flow cytometry. As shown in Figure 13B, applying a gate can further improve the identification rate of each event.

[0149] Based on the above, by clarifying the relationship between secretory cells (especially antibody-expressing cells), expressed secretions (especially expressed antibodies), and antigen-carrying particles (especially cells that present antigens to which expressed antibodies specifically bind) using positional information, it becomes possible to detect the phenomenon of interest (i.e., the state in which the secretions are bound to molecules possessed by the antigen-carrying particles). Furthermore, by quantifying the degree of this phenomenon of interest, the required level of accuracy can be adjusted for each user, allowing for a determination of how reliable the cell population should be before passing it on to subsequent processes. As a result, the cost of screening (especially antibody screening) can be controlled from the user's perspective.

[0150] In the binding state detection step S107, by also considering the difference in fluorescence intensity, information about the phenomenon of interest described above can be obtained with higher accuracy. Figure 14 shows an overview of the case where the difference in fluorescence intensity is considered. The leftmost figure in the top row of Figure 14 shows a bright-field image, the figure next to it shows a fluorescence image of secretory cells (Donor; D), the figure next to that shows a fluorescence image of antigen-holding particles (Receiver; R), and the figure next to that shows a fluorescence image of secretion (Secretion; S). Each fluorescence image can be obtained using a separate fluorescence channel. From these images, as shown in the middle row of Figure 14, mask information for secretory cells and antigen-holding particles is extracted, and the fluorescence intensity at the masked position is calculated from these fluorescence signals. In the case shown in Figure 14, as shown in the rightmost figure in the middle row of Figure 14, some of the fluorescence intensities overlap. Therefore, it is necessary to separate the fluorescence intensity from secretory cells from the fluorescence intensity from antigen-containing particles using some method.

[0151] Furthermore, we will explain using Figure 15, which is a one-dimensional representation of the diagram shown in Figure 14. Figure 15A shows the fluorescence signal from secretions originating from secretory cells. The left diagram (Case 1) shows the case where both the secretory cells and the antigen-containing particles have similar fluorescence signals, while the right diagram (Case 2) shows the case where the fluorescence signal from the antigen-containing particles is small, i.e., where there is little secretion originating from the antigen-containing particles. Figure 15B shows the fluorescence signal from secretions originating from antigen-containing particles. The left and right diagrams are the same as those described in Figure 15A. Figure 15C shows the fluorescence signal of secretions actually observed. The left and right diagrams are the same as those described in Figure 15A.

[0152] As shown in Figure 15, even if the fluorescence signal from secretions originating from antigen-containing particles is weak, the overlap rate alone will result in the same judgment in both Case 1 and Case 2. In this case, simply dividing the fluorescence signal in half, for example, is insufficient, and the fluorescence signal from secretions originating from antigen-containing particles will be incorrectly calculated as being higher than it actually is. Therefore, in Case 2, it may actually be better to judge it as "negative."

[0153] In contrast, Figure 16 illustrates a method for calculating the overlap of each fluorescence signal. As shown in the upper left of Figure 16, in the fluorescence signals from the fluorescence channels of secretory cells and antigen-containing particles, the fluorescence signals of the secretory cells and the antigen-containing particles partially overlap in interval Z. In this case, as shown in the lower left of Figure 16, the overlapping interval Z is summed and detected in the fluorescence signals of the secretions. On the other hand, information that can only be obtained from secretory cells, such as the area Ad1 at time Wd1, and information that can only be obtained from antigen-containing particles, such as the area Ar2 at time Wr2, can be obtained as information. In addition, the summed area Total(Ad2+Ar1) in interval Z can also be obtained as information. Therefore, by using this information and Wd2 or Wr1 corresponding to the interval Z time, the area of ​​the other unknown parts (Ad2, Ar1) can be calculated.

[0154] Ad2 and Ar1 can be calculated from the ratio, for example, as shown in formulas (2) and (3) below.

[0155]

[0156] Furthermore, Ad2 and Ar1 can be calculated, for example, by squaring the area assuming it follows a normal distribution, as shown in equations (4) and (5) below.

[0157]

[0158] Furthermore, since Ad2 and Ar1 obtained using equations (2) to (5) above are estimated values, they do not necessarily match Total. Therefore, as shown in equations (6) to (8) below, by distributing Total in the ratio of Ad2:Ar1, we can obtain a more accurate Ad2 (=Ad2') and a more accurate Ar1 (=Ar1').

[0159]

[0160] In the specific example described above, a one-dimensional calculation method is used, utilizing the known fluorescence signal (particularly the area). However, the basic concept is the same for a two-dimensional calculation method. Furthermore, while Figure 15 shows a case where the fluorescence signal from secretions derived from antigen-containing particles is weak, it is also possible that the fluorescence signal from secretions derived from secretory cells may be weak. In this case as well, by applying the specific example described above, it is possible to calculate the fluorescence signal while considering the difference in fluorescence intensity.

[0161] (2-8) Other processes

[0162] The secretion evaluation method according to this embodiment may further include other steps. An example of such other steps is described below.

[0163] <Culture process>

[0164] In this embodiment, when an emulsion is used as a carrier, in which the culture medium is the dispersed phase and the liquid is immiscible with the culture medium as the dispersion medium, secretory cells and / or antigen-bearing particles can be cultured within the carrier. The type of culture medium can be appropriately selected depending on the secretory cells and / or antigen-bearing particles to be cultured.

[0165] <Carrier destruction process>

[0166] In this embodiment, the carrier may be destroyed as necessary after the preparation step S106, etc. In another embodiment, each secretory cell and / or antigen-containing particle within the carrier may be destroyed. In this case, the destruction may be carried out while maintaining the carrier. This allows, for example, components derived from cells or particles to be released into the carrier, and these components can be processed separately from other components.

[0167] <Detection Process>

[0168] In this embodiment, after the preparation step S106, the components of the carrier may be detected, analyzed, or reacted with other components. For example, the carrier may be merged with other carriers in order to perform the detection or reaction. After the merging, the components of the carrier may be detected, analyzed, or reacted with other components. The carrier may also react with a drug. This makes it possible to detect or analyze the carrier's response to the drug.

[0169] <Synthesis process>

[0170] In this embodiment, after the preparation step S106, etc., the synthesis of chemical substances by secretory cells and / or antigen-bearing particles in the carrier may be carried out. For example, this synthesis may be carried out in the carrier. For example, by encapsulating a cell-free expression reagent in the carrier, it becomes possible to produce an in vitro antibody. Examples of the cell-free expression reagent include linear DNA, promega (E. coli S30 Extract system for linear DNA), etc.

[0171] (3) Specific examples of gate representation

[0172] A specific example of gate display will be described below. Note that this display may be performed via an output unit (e.g., a display, monitor, portable information terminal, wearable terminal, etc.) of the information processing unit 103 and / or processing unit 200 of the particle analyzer 100, but is not limited to this in this embodiment.

[0173] (3-1) Specific Example 1

[0174] Figure 18 shows a specific example of gate display 1. In this specific example 1, secretory cells (donor cells) are first stained with a cell staining reagent (e.g., Calcein AM), and antigen-expressing cells (receiver cells) contained in antigen-holding particles are stained with a cell staining reagent (e.g., Calcein Blue). These cells are then introduced into a carrier and measured using a flow cytometer. Next, a graph is created using the staining signals from secretory cells and antigen-expressing cells. Populations where the staining signals from both secretory cells and antigen-expressing cells are high are then gated as regions where both secretory cells and antigen-expressing cells are present in the carrier (i.e., the heterocellular regions shown in Figure 18).

[0175] In the gate display shown in Figure 18, the horizontal axis represents D cell (fluorescence intensity derived from staining of secretory cells), and the vertical axis represents R cell (fluorescence intensity derived from staining of antigen-expressing cells). In Figure 18, the upper left can be divided into a region containing only antigen-expressing cells, the upper right into a heterozygous cell region (gate region), the lower left into a region without cells or particles, and the lower right into a region containing only secretory cells. The setting of the gating (threshold) can include, for example, the maximum value of the negative control, the quartile, the top few percent of the negative control, the top few percent of the target sample, or a value arbitrarily defined by the user, but is not limited to these in this embodiment.

[0176] (3-2) Specific Example 2

[0177] Figure 19 shows a specific example of gate display 2. In this specific example 2, first, secretory cells (donor cells) are stained with a cell staining reagent (e.g., Calcein AM), and antigen-expressing cells (receiver cells) contained in antigen-holding particles are stained with a cell staining reagent (e.g., Calcein Blue). These cells are then introduced into a carrier and measured using a flow cytometer. Next, a graph is created using the staining signals from secretory cells and antigen-expressing cells. Populations where the staining signals from both secretory cells and antigen-expressing cells are high are then gated as regions where both secretory cells and antigen-expressing cells are present in the carrier (i.e., heterocellular regions).

[0178] Next, the gated heterocellular region is evaluated based on the fluorescence signal intensity derived from secretions (especially antibodies). For example, a scatter plot can be used to show both the secretion signal and the antigen-expressing cell signal. In addition to the secretion signal, other secretory cell signals (e.g., D ab (fluorescence intensity derived from secretions of secretions from secretory cells)) and light scattering may also be used. Alternatively, instead of a scatter plot, a histogram or bar graph based on frequency of occurrence can be used. Regions with high R ab (fluorescence intensity derived from secretions of antigen-expressing cells) can be defined as high-binding secretion signal regions (see Figure 19). Note that the setting of the gating (threshold) can be done in the same way as in Specific Example 1 described above, so the explanation is omitted here.

[0179] (3-3) Specific Example 3

[0180] Figure 20 shows a specific example of gate display 3. In this specific example 3, first, secretory cells (donor cells) are stained with a cell staining reagent (e.g., Calcein AM), and antigen-expressing cells (receiver cells) contained in antigen-carrying particles are stained with a cell staining reagent (e.g., Calcein Blue). These cells are then introduced into a carrier and measured using a flow cytometer. Next, a graph is created using the staining signals from secretory cells and antigen-expressing cells. Populations where the staining signals from both secretory cells and antigen-expressing cells are high are then gated as regions where both secretory cells and antigen-expressing cells are present in the carrier (i.e., heterocellular regions).

[0181] Next, the gated heterocellular region is evaluated based on the fluorescence signal intensity derived from secretions (especially antibodies). For example, a scatter plot can be used to show both the secretion signal and the antigen-expressing cell signal. In addition to the secretion signal, other methods such as secretory cell signals and light scattering can be used. Alternatively, instead of a scatter plot, histograms or bar graphs based on frequency of occurrence can be used. Regions with high R ab (fluorescence intensity derived from secretions of antigen-expressing cells) can be defined as high-binding secretion signal regions. Alternatively, instead of cell staining signals, Cr ab (fluorescence intensity derived from secretions on the carrier), Cr' ab (fluorescence intensity derived from secretions in the region Cr' = Cr-D-R, which excludes the D and R regions from the Cr region; Cr' ab = Cr ab - D ab (fluorescence intensity derived from secretions of secretions of secretions of secretions of secretions of antigen-expressing cells)), etc., can be used. Furthermore, in Figure 20, by appropriately setting the gating, the region can be further divided into a high secretion, high binding signal region and a low secretion, high binding signal region. In addition, instead of a scatter plot, for example, a histogram or bar graph based on frequency of occurrence can be used. Note that the gating (threshold) can be set in the same way as in Specific Example 1 described above, so the explanation is omitted here.

[0182] (3-4) Specific Example 4

[0183] Figure 21 shows a specific example of gate display 4. In this specific example 4, first, secretory cells (donor cells) are stained with a cell staining reagent (e.g., Calcein AM), and antigen-expressing cells (receiver cells) contained in antigen-carrying particles are stained with a cell staining reagent (e.g., Calcein Blue). These cells are then introduced into a carrier and measured using a flow cytometer. Next, a graph is created using the staining signals from secretory cells and antigen-expressing cells. Populations where the staining signals from both secretory cells and antigen-expressing cells are high are then gated as regions where both secretory cells and antigen-expressing cells are present in the carrier (i.e., heterocellular regions).

[0184] Next, the gated heterocellular region is evaluated based on the fluorescence signal intensity derived from secretions (especially antibodies). For example, a scatter plot can be used to show both the secretion signal and the antigen-expressing cell signal. In addition to the secretion signal, other methods such as secretory cell signals and light scattering can be used. Alternatively, instead of a scatter plot, histograms or bar graphs based on frequency of occurrence can be used. Regions with high R ab (fluorescence intensity derived from secretions of antigen-expressing cells) can be defined as high-binding secretion signal regions. Alternatively, instead of cell staining signals, Cr ab (fluorescence intensity derived from secretions on the carrier), Cr' ab (fluorescence intensity derived from secretions in the region Cr' = Cr-D-R, which is the region obtained by removing the D and R regions from the Cr region; Cr' ab = Cr ab - D ab (fluorescence intensity derived from secretions of secretions of secretions of secretions of secretions of secretions of antigen-expressing cells)), etc., can be used. R ab / Cr' ab can also be used as a specific binding index. As shown in Figure 21, regions with high specific binding indices can be gated as high-affinity secretion regions. Furthermore, instead of a scatter plot, histograms or bar graphs based on frequency of occurrence can be used, for example. Note that the gating (threshold) can be set in the same way as in Specific Example 1 described above, so the explanation is omitted here.

[0185] (3-5) Specific Example 5

[0186] Figure 22 shows a specific example of gate display 5. In this specific example 5, first, secretory cells (donor cells) are stained with a cell staining reagent (e.g., Calcein AM), and antigen-expressing cells (recipient cells) contained in antigen-holding particles are stained with a cell staining reagent (e.g., Calcein Blue). These cells are then introduced into a carrier and measured using a flow cytometer. Next, a graph is created using the staining signals from secretory cells and antigen-expressing cells. Populations where the staining signals from both secretory cells and antigen-expressing cells are high are then gated as regions where both secretory cells and antigen-expressing cells are present in the carrier (i.e., heterocellular regions).

[0187] Next, the gated heterocellular region is evaluated based on the fluorescence signal intensity derived from secretions (especially antibodies). For example, secretion signals and antigen-expressing cell signals can be used in a scatter plot. In addition to secretion signals, secretory cell signals and light scattering can also be used. Alternatively, instead of a scatter plot, histograms and bar graphs based on frequency of occurrence can be used. Regions with high R ab (fluorescence intensity derived from secretions of antigen-expressing cells) can be defined as high-binding secretion signal regions. Alternatively, instead of cell staining signals, Cr ab (fluorescence intensity derived from secretions on the carrier), Cr' ab (fluorescence intensity derived from secretions in the region Cr' = Cr-D-R, which is the region obtained by removing the D and R regions from the Cr region; Cr' ab = Cr ab - D ab (fluorescence intensity derived from secretions of secretions of secretions of secretions of secretions of secretions of antigen-expressing cells)), etc., can be used. Alternatively, instead of a scatter plot, histograms and bar graphs based on frequency of occurrence can be used.

[0188] Furthermore, by using E Ag (fluorescence intensity derived from antigens in antigen-expressing cells), the dissociation constant index Kd' can be calculated based on the following formula (9) instead of the specific binding index R ab / Cr' ab, and as shown in Figure 22, regions with high Kd' can be defined as high-affinity secretion regions.

[0189]

[0190] Note that the gating (threshold) settings can be configured in the same way as in Specific Example 1 described above, so we will omit the explanation here.

[0191] (4) Specific examples of presenting the true positive rate

[0192] Ideally, we want to obtain an ab fluorescence signal only when true secretory cells that secrete a substance (in this case, an antibody) against the target antigen contained in the antigen-carrying particle are present, along with true antigen-carrying particles. However, when non-positive (false-positive) secretory cells that secrete substances against antigens other than the target antigen are present, along with antigen-carrying particles, an ab fluorescence signal may occur in some cases. For example, secretions from true secretory cells may enter the carrier containing non-positive secretory cells.

[0193] Here, "positive" consists of true secretory cells and non-positive secretory cells. However, the identification of true positives and non-positives may be based on verification by other processes performed later, making it difficult to distinguish between true positives and non-positives. In particular, in antibody screening, the number and proportion of non-positives are often greater than that of true positives. As a result, problems arise such as true positives being buried among non-positives and not being discovered, or the number of subsequent processes increasing, wasting time and increasing costs.

[0194] In contrast, by detecting ab fluorescence signals in populations without secretory cells and calculating their proportion, it is possible to predict the true positive and false positives of ab fluorescence signals in populations with secretory cells. Furthermore, by calculating the false positive rate, the proportion of true positives among positives can be calculated, and software to display these figures can also be provided. In addition, it is possible to predict ab fluorescence signals in populations with secretory cells from the proportion of populations without ab fluorescence signals.

[0195] Figure 23 shows the assumptions in a specific example of presenting the true positive rate. Here, it is assumed that the probability of an ab fluorescence signal being present in a carrier without secretory cells is equivalent to the probability of an ab fluorescence signal being present (without the involvement of true positive secretory cells) in a carrier with secretory cells.

[0196] Figure 24 shows an overview of examples of determining the true positive rate. Figure 24A shows the positive group (population containing secretory cells) determined to have secretory cells (especially antibody-producing cells), divided into cases where one cell is present in the carrier and cases where multiple cells are present. Figure 24B shows the group (population without secretory cells) determined to have no secretory cells, divided into cases where no cells are present in the carrier and cases where antigen-carrying particles are present.

[0197] The patterns shown in each figure are explained below. In Figure 24A, in the case of a single cell, there are cases where a truly positive secretory cell is in the carrier and cases where a non-positive secretory cell is in the carrier. In the case of multiple cells, there are cases where a truly positive secretory cell and antigen-carrying particles with secretions bound to the target antigen are in the carrier, and cases where a non-positive secretory cell and antigen-carrying particles with secretions bound to the antigen are in the carrier. In Figure 24B, there are cases where no secretory cells are in the carrier and an ab fluorescence signal is present, and cases where no secretory cells are in the carrier and there is no ab fluorescence signal. Furthermore, there are cases where no secretory cells are in the carrier but antigen-carrying particles with secretions bound to the target antigen are in the carrier and there is an ab fluorescence signal, and cases where no secretory cells are in the carrier but antigen-carrying particles without bound secretions are in the carrier and there is no ab binding signal. Here,

[0198]

[0199] If we divide them into categories, for example, since 10% are positive secretory cells, the probability of a positive secretory cell and the probability of a non-positive secretory cell can be expressed as shown in the following formula (10).

[0200]

[0201] Next, if we assume that the probability of fluorescence being present in non-positive secretory cell carriers is the same as the probability of fluorescence being present in empty carriers, then this can be expressed as shown in the following formula (11).

[0202]

[0203] The probability of a non-positive secretory cell carrier exhibiting fluorescence, and the probability of a positive secretory cell carrier exhibiting fluorescence, can be calculated using the following formulas (12) and (13), respectively.

[0204]

[0205] Therefore, the probability that a cell is positive for secretory cells when fluorescence is present, and the probability that a cell is not positive for secretory cells when fluorescence is present, can be calculated using the following formulas (14) and (15), respectively.

[0206]

[0207] Figure 25 shows a specific example of the presentation of the true positive rate. In Figure 25, the number of positive secretory cells is 50, the percentage of positive secretory cells is 10%, and among the positives, the true positives are represented as 25 true positive secretory cells and 50% (or 5% of the total), with the highly bound secretory signaling region being gated. Here, examples of displaying the gated population include displaying one or more selected from the group consisting of the number of positives (number of positive secretory cells), the percentage of positives (percentage of positive secretory cells), the number of true positives (number of true positive secretory cells), and the percentage of true positives (percentage of true positive secretory cells). In this embodiment, it is particularly preferable to display the number of true positives and / or the percentage of true positives. Note that this embodiment also includes a form in which the number of positives and the percentage of positives are not included in the display, but one or more of the number of true positives and the percentage of true positives are displayed.

[0208] 2. Second Embodiment (Secretion Evaluation System 1000)

[0209] (1) Description of the second embodiment

[0210] Figure 17 schematically shows the overall configuration of the secretion evaluation system 1000 according to this embodiment. The secretion evaluation system 1000 according to this embodiment includes a particle analyzer 100 which comprises at least a flow path C through which a liquid containing particles flows, an irradiation unit 101 which irradiates light onto particles flowing through the flow path C, and a detection unit 102 which detects the light generated by the light irradiation; and a processing device 200 which performs at least a position information identification step S101 which identifies the position information of the secretion cell and the antigen-containing particle based on optical information relating to the secretion cell and the carrier holding at least one antigen-containing particle, a secretion information acquisition step S102 which acquires information on secretions that bind to unbound molecules possessed by the secretion cell and the antigen-containing particle, and a secretion evaluation step S103 which evaluates the characteristics of the secretion based on the position information and the information on the secretion.

[0211] (2) An example of a second embodiment

[0212] In the secretion evaluation system 1000 according to this embodiment, the particle analyzer 100 and the processing unit 200 may reside in a single device, or they may be distributed across multiple devices. Furthermore, the particle analyzer 100 and the processing unit 200 may be connected to a network wirelessly or via a wired connection. In addition, the processing unit 200 may be configured to be able to store and execute programs, for example, in the information processing unit 103 of the particle analyzer 100, on the internet, or in the cloud. Each device will be described below.

[0213] (2-1) Particle analyzer 100

[0214] In this embodiment, the particle analyzer 100 may optionally include a sorting unit 104 for sorting particles (particularly cells, and more particularly secretory cells) that have been determined to be sorted. The irradiation unit 101, detection unit 102, information processing unit 103, and sorting unit 104 may be located in a single device, or they may be distributed and mounted in multiple devices.

[0215] The particle analyzer 100 can specifically include a flow cytometer and an imaging cytometer. Furthermore, the particle analyzer 100 including the sorting unit 104 is specifically referred to as a cell sorter.

[0216] The flow path C, irradiation unit 101, detection unit 102, information processing unit 103, and sorting unit 104 are the same as those described in "(2) An Example of the First Embodiment" of "1. First Embodiment (Method for Evaluating Secretions)," so their explanation is omitted here.

[0217] The microchip described above may be used in the particle analyzer 100 according to this embodiment and may be removable from the particle analyzer 100. By making the microchip removable from the particle analyzer 100, a new microchip can be used for each sample S, thereby preventing contamination.

[0218] (2-2) Processing apparatus 200

[0219] In this embodiment, the processing apparatus 200 may, if necessary, perform one or more steps selected from the group consisting of the photodetection step S104, the capture molecule placement step S105, the separation step S106, and the binding state detection step S107 described above.

[0220] The location information identification step S101, secretion information acquisition step S102, secretion evaluation step S103, photodetection step S104, capture molecule placement step S105, sorting step S106, and binding state detection step S107 are the same as those described in "(2) An Example of the First Embodiment" of "1. First Embodiment (Secretion Evaluation Method)," so their explanation is omitted here.

[0221] In this embodiment, the processing unit 200 may include an output unit (e.g., a display, monitor, portable information terminal, wearable terminal, etc.) or an input unit (e.g., a keyboard, voice input device, etc.) for performing output or input, similar to the information processing unit 103.

[0222] Furthermore, the following configurations may also be adopted in this technology: [1] A method for evaluating secretions, comprising: a position information identification step of identifying the position information of a secretory cell and an antigen-holding particle based on optical information relating to a carrier holding at least one secretory cell and at least one antigen-holding particle; a secretion information acquisition step of acquiring information relating to secretions that bind to unbinding molecules held by the secretory cell and the antigen-holding particle; and a secretion evaluation step of evaluating the characteristics of the secretions based on the position information and the information relating to the secretions. [2] The method for evaluating secretions according to [1], further comprising a photodetection step of detecting light from the carrier and acquiring optical information relating to the carrier. [3] The method for evaluating secretions according to [1] or [2], wherein the optical information relating to the carrier is an image or fluorescence information. [4] The secretion evaluation method according to any one of [1] to [3], wherein in the secretion information acquisition step, the difference in signals between unstained secretions and stained secretions, or the signal ratio of unstained secretions and stained secretions, on the secretory cells and the antigen-holding particles is acquired as information regarding secretions that bind to the unbinding molecules held by the secretory cells and the antigen-holding particles. [5] The secretion evaluation method according to [4], wherein in the secretion information acquisition step, an evaluation index consisting of subtraction or division of unstained secretions is further used. [6] The secretion evaluation method according to any one of [1] to [5], further comprising a capture molecule placement step of placing capture molecules that capture the secretions on the carrier. [7] The secretion evaluation method according to [6], wherein in the capture molecule placement step, the amount of secretions that are unbound to the unbinding molecules is detected. [8] The secretion evaluation method according to [6] or [7], wherein, in the capture molecule placement step, the antigen-containing particles are made to coexist in the carrier in advance, and the amount of secretion that shows nonspecific binding to the untetherable molecule is measured. [9] The secretion evaluation method according to any one of [6] to [8], wherein, in the capture molecule placement step, the capture molecules are immobilized in the carrier in advance, and the amount of secretion that is not bound to the untetherable molecule is measured.

[10] A method for evaluating secretions according to any one of [1] to [9], wherein in the position information identification step, the carrier further holds at least one non-antigen particle, and the position information of the non-antigen particle is further identified based on optical information relating to the carrier, and the position information of the non-antigen particle is used in the secretion information acquisition step.

[11] A method for evaluating secretions according to

[10] , wherein in the secretion information acquisition step, the difference in signals between the non-antigen particle and the antigen-holding particle, or the signal ratio of the non-antigen-holding particle and the antigen-holding particle is acquired as information relating to secretions that bind to the binding molecule held by the secretory cell and the antigen-holding particle.

[12] A method for evaluating secretions according to [2], wherein in the photodetection step, fluorescence from the fluorescently labeled carrier is detected, and in the position information identification step, the optical information relating to the carrier is corrected based on the detected fluorescence intensity of the carrier.

[13] The secretion evaluation method according to

[12] , wherein in the position information identification step, a fluorescence correction coefficient is calculated from the fluorescence intensity of the detected carrier, and one or more fluorescence intensity selected from the group consisting of the secretory cells, the antigen-holding particles, the secretion, and the carrier surface is corrected.

[14] The secretion evaluation method according to any one of [1] to

[13] , wherein the characteristic is the binding affinity or binding specificity between the unbinding molecule held by the antigen-holding particles and the secretion.

[15] The secretion evaluation method according to

[14] , further comprising a sorting step of sorting the secretory cells if the binding specificity or binding affinity is above a certain level.

[16] The secretion evaluation method according to any one of [1] to

[15] , further comprising a binding state detection step of further identifying the position information of the secretion in the position information identification step, and detecting the state in which the secretion is bound to the unbinding molecule held by the antigen-holding particles based on the relationship between the position information of the secretory cells, the antigen-holding particles, and the secretion.

[17] The secretion evaluation method according to

[16] , wherein in the binding state detection step, the degree to which the secretion is bound to the unbound molecule possessed by the antigen-possessing particle is quantified based on the fluorescence signals of the secretory cell, the antigen-possessing particle, and the secretion.

[18] The secretion evaluation method according to

[17] , wherein the quantification uses the overlap of each of the fluorescent signals.

[19] The secretion evaluation method according to any one of [1] to

[18] , wherein in the secretion evaluation step, a signal that changes when the secretion binds to the unbinding molecule possessed by the antigen-possessing particle is detected, and the detected signal is used for the evaluation.

[20] A particle analyzer comprising: a flow channel through which a liquid containing particles flows; an irradiation unit that irradiates light onto particles flowing through the flow channel; and a detection unit that detects the light generated by the light irradiation. A processing device that performs: a position information identification step that identifies the position information of the secretion cell and the antigen-possessing particle based on optical information relating to at least one secretion cell and a carrier holding at least one antigen-possessing particle; a secretion information acquisition step that acquires information on a secretion that binds to the unbinding molecule possessed by the secretion cell and the antigen-possessing particle; and a secretion evaluation step that evaluates the characteristics of the secretion based on the position information and the information on the secretion.

[0223] The present technology will be described in more detail below based on the following examples. The examples described below are representative examples of the present technology and should not be interpreted as narrowing the scope of the present technology.

[0224] <Example 1>

[0225] This example describes a test system for searching for antibodies or ligands (secretions) that bind to GPCR receptors (antigens) expressed by GPCR (G Protein-Coupled Receptor) stably expressing cells (antigen-carrying particles). In this example, CCR5 (C-C chemokine receptor type 5) was selected as an example of a GPCR.

[0226] Table 1 below summarizes the cells or antibodies, cells, ligands or receptors, cell staining conditions, ligand-recognizing antibodies, and functional assays.

[0227]

[0228] Figure 26A shows the reaction between cells and secretions. In Figure 26, antibody or ligand-producing cells are designated as Secretion cells (S), and GPCR-stable expression cells (particularly CCR5 receptor-expressing cells) are designated as Receiver cells (R).

[0229] Examples of GPCR-stable expression cells include CCR5-stable expression CHO cells, β2-adrenergic receptor (β2AR)-stable expression cells in HEK293 cells, histamine H1 receptor-stable expression cells in CHO cells, cannabinoid CB1 receptor-stable expression cells in U2OS cells, muscarinic M2 receptor-stable expression cells in CHO cells, and serotonin 5-HT2A receptor-stable expression cells in HT29 cells, but are not limited to these in this embodiment.

[0230] Examples of CCR5 receptor-expressing cells include the CCR5 stable-expressing CHO cell line. In this case, CCR5-non-expressing CHO cells can also be used as a negative control in this embodiment. Furthermore, a CCR5-expressing TEV protease assay is known, and this assay can also evaluate the expression and function of a specific cell surface receptor called CCR5 using TEV (tobacco etch virus) protease.

[0231] Figure 26B shows patterns of signal detection methods. In pattern (i), when an antibody or ligand is labeled using a fluorescently labeled antibody, the fluorescent label of the antibody / ligand bound to the GPCR expressed by GPCR-stable expression cells is detected, for example, using the detection unit 102, as shown in Figure 26B.

[0232] On the other hand, pattern (ii) involves performing a functional assay to detect changes in the fluorescence signal within GPCR-stable expressing cells associated with antibodies / ligands bound to GPCRs.

[0233] In this example, the calcium assay can be particularly useful as an indicator for evaluating CCR5 signal activation. CCR5 signal activation is associated with the activation of calcium ions (Ca) in the cytoplasm. ２＋It is known that the concentration increases. In fact, there is data showing that when CHO-K1 is stimulated with ATP, the calcium ion concentration increases and the fluorescence intensity increases. Furthermore, since it is known that the calcium ion concentration in the cytoplasm also increases with the addition of ATP, it is thought that it can be used as a positive control.

[0234] Therefore, activation of the CCR5 signaling pathway increases the intracytoplasmic calcium ion concentration, and by detecting the changes in fluorescence intensity and fluorescence spectrum in response to the calcium ion concentration, it can be used as an indicator for the aforementioned evaluation.

[0235] In addition, reporter assays can also serve as indicators for evaluating CCR5 signaling activation. When a ligand binds and the receptor is activated, protease-labeled β-arrestin is recruited to a Cterminus-modified GPCR containing a transcription factor bound to the protease cleavage site. The protease then cleaves the transcription factor from the GPCR, which immediately translocates to the nucleus, activating β-lactamase activity. When a substrate is loaded, if β-lactamase is not expressed, the cells emit green fluorescence. If β-lactamase is expressed, the substrate is cleaved, and the cells emit blue fluorescence.

[0236] Therefore, activation of the CCR5 signaling pathway activates β-lactamase activity, and when β-lactamase is expressed, the substrate is cleaved, and the cells emit blue fluorescence, which can be detected and used as an indicator for the aforementioned evaluation.

[0237] Furthermore, GPCR internalization can also be used as an evaluation indicator. Examples of GPCR internalization assays include Activated Internalization assays and Total Endocytosis assays.

[0238] In the Activated Internalization assay, an enzyme fragment localized to the endosome surface of cells is used, along with a complementary enzyme fragment fused to β-arrestin. β-arrestin binds to GPCRs upon stimulation of the target receptor, leading to receptor internalization and transport to endosomes. As a result, the two enzyme fragments complement each other, and the increase in enzyme activity can be measured using a chemiluminescent detection reagent. The results of this measurement can be used as an indicator for the aforementioned evaluation.

[0239] The Total Endocytosis assay uses an enzyme fragment localized on the surface of cell endosomes and a complementary enzyme fragment attached to a target receptor. Stimulation by a ligand or agonist leads to the uptake of the target receptor into the cell and transport to the endosome. As a result, the two enzyme fragments complement each other independently of β-arrestin, and the increase in enzyme activity can be measured using a chemiluminescent detection reagent. The results of this measurement can be used as an indicator for the aforementioned evaluation.

[0240] In addition, the cAMP assay can also serve as an evaluation indicator. Examples of cAMP assays include the HitHunter® cAMP assay (manufactured by DiscoverX). This cAMP assay is a competitive immunoassay, specifically, in which the active β-galactosidase enzyme is fragmented into two, one of which is labeled with cAMP, and the free cAMP in the cell lysate is made to compete with the labeled cAMP. The labeled cAMP that is not bound to the antibody then spatially approaches the other fragment of the enzyme contained in the reagent and is reconstituted as active β-galactosidase. As a result, the enzymatic activity of the reconstituted β-galactosidase hydrolyzes the chemiluminescent substrate, and the resulting chemiluminescent signal can be detected for measurement. The results of this measurement can be used as an evaluation indicator.

[0241] <Example 2>

[0242] This embodiment describes automatic gate selection.

[0243] Figure 28 illustrates the process of gate selection by the user. Gate selection consists of several steps, but at least the following steps are required. First, as a preliminary step, gates with one carrier are gated as shown in Figure 28. Next, as shown in Figure 28A, heterologous (Donor+, Receiver+) gates are selected, in which the carrier contains both secretory cells (Donor cells) and antigen-expressing cells (Receiver cells) contained in the antigen-holding particles. Specifically, the gate shown in Figure 28A is divided into approximately four quadrants, and the heterologous gate with the highest brightness appears in the first quadrant, so that part is gated. In this case, the second quadrant shows gates in which the carrier contains only antigen-expressing cells, the third quadrant shows gates in which the carrier contains neither type of cell, and the fourth quadrant shows gates in which the carrier contains only secretory cells.

[0244] Next, a child gate is selected from the gates selected in Figure 28A. Specifically, among the gates selected in Figure 28A, there are some that are heterogeneous but do not have a signal from the antibody. Therefore, as shown in Figure 28B, a gate that contains a Secretion (signal from the antibody) (Secretion+) is selected as the child gate.

[0245] Finally, a child gate is selected from the gates selected in Figure 28B. Specifically, as shown in Figure 28C, a scattergram is created using the Donor overlap and Receiver overlap indicators. This will result in a Receiver with a Secret appearing in the upper right, allowing the user to gate to the desired state carrier (i.e., Hetero with a Receiver with a Secret) by drawing a roughly triangular gate.

[0246] Here, the user's gate selection is based on the method described above, but in the gate selection shown in Figure 28B and Figure 28C, it is not clear to the user where to gate, making it difficult to perform optimal gating.

[0247] In contrast, this embodiment proposes a method for automatically selecting gates in the selection of gates shown in Figure 28B and Figure 28C.

[0248] Figure 29 is a diagram illustrating the overview of automatic gate selection. In automatic gate selection, first, the gating shown in Figure 28B and Figure 28C are varied (see Figures 29A and 29B). Specifically, in the graph on the left of Figure 29B, the threshold of Peak intensity (Secretion) is varied, and in the graph on the right of Figure 29B, the roughly triangular gate is changed to various sizes. Next, as shown in Figure 29C, feature quantities are extracted from the image group contained in the gate at that time. Finally, using the feature quantities obtained when the threshold and the roughly triangular gate are varied in the two graphs shown in Figure 29B, the appropriate gate is determined.

[0249] In this embodiment, the method using one feature and the method using two features will be described in detail below.

[0250] [Method using a single feature]

[0251] In this method, the feature vector used is "the number of samples in which the area of ​​a pixel containing both Receiver and Secretion is greater than or equal to a certain value." For example, suppose there are 20 samples in the gate shown on the left side of B in Figure 29 and the gate shown on the right side of B in Figure 29. Next, we calculate the area of ​​each sample that is both Receiver+ and Secretion+ (the target area; see the light gray area in Figure 30). Suppose there are 18 samples in which this area is 17 or less. Then, we use this number of samples, 18, as the feature vector.

[0252] Furthermore, the presence of both Receiver+ and Secretion+ in a certain area indicates a high probability of a "positive" sample. Therefore, the proportion calculated here can be said to represent the proportion of gates shown on the left side of Figure 29B and gates shown on the right side of Figure 29B that are likely to contain "positive" samples.

[0253] Figure 31 is a flowchart showing an example of an embodiment of a method that utilizes a single feature. The following processing is, in principle, performed in the information processing unit 103. In this embodiment, first, the threshold (for example, six thresholds are set in advance) is set to a high value at the gate shown in A of Figure 31 (S201). Next, the threshold set at the gate shown in A of Figure 31 is applied to all the gates shown in B of Figure 31 (S202). Specifically, all gates from 0 to 8 shown in B of Figure 31 are applied, and the feature quantities are calculated for each obtained image. Next, the feature quantities of the sample group are calculated (S203). Next, the gate that is roughly triangular in B of Figure 31 is selected, where "number of matches to the feature" / "total number is X% or more" (S204). Here, the value of X can be set to any number, such as 85%.

[0254] Next, it is determined whether the predetermined threshold has been reached (S205). If it is determined that the predetermined threshold has been reached, the conditions under which the "number of matches to the features" increases sharply are sought, as will be described later (S206). On the other hand, if it is determined that the predetermined threshold has not been reached, the threshold is lowered (S207), and the process returns to S202.

[0255] The method for finding the conditions under which the "number of matches to the above-mentioned features" increases sharply is explained below.

[0256] Figure 32 illustrates a method for finding the conditions under which the number of features matches sharply increases. In Figure 32A, the vertical axis represents the number of samples that match the features, and the horizontal axis represents the threshold. As shown in Figure 32A, it can be seen that there is a point where the number of features matches sharply increases as the threshold is lowered. Based on experience, unintended samples enter the model at this point of sharp increase, so finding this point of sharp increase improves performance. Therefore, we find the threshold corresponding to this point of sharp increase.

[0257] To find the points of sharp increases mentioned above, we use the derivative values ​​of a number that match the characteristics, as shown in Figure 32B. The dotted line in Figure 32B represents the derivative values. The solid line in Figure 32C shows the approximation curve of the derivative values ​​so far, where "○" represents the next actual derivative value and "×" represents twice the estimated value obtained from the approximation curve. The specific method for finding points of sharp increases using these graphs will be explained below.

[0258] Figure 33 is a flowchart showing an example of a specific method for identifying points of rapid increase. The following processing is, in principle, performed in the information processing unit 103. Figures 33A to F are graphs that partially and chronologically show what happens when the flowchart shown in Figure 33G is repeatedly performed.

[0259] In this embodiment, first, assuming that we start with n=10, we extract the first n data points (S301). Next, we calculate the moving average of the 10 points in the window (S302). Then, we calculate the derivative of the above and calculate the moving average of the 10 points (S303). Note that the dotted lines A to F in Figure 33 represent the moving average of the derivative. Next, we obtain the above linear approximation formula (S304). Note that the solid lines A to F in Figure 33 represent the linear approximation formula. Next, we determine whether the slope of the linear approximation formula is ≥0.001 (S305).

[0260] If the slope of the linear approximation formula is determined to be ≥ 0.001, then the estimated values ​​are calculated from the linear approximation formula for the next three points, and it is determined whether the measured values ​​are twice or more the estimated values ​​(S306). If it is determined that the measured values ​​are twice or more the estimated values, the point of rapid increase is determined, and the process is terminated. On the other hand, if it is determined that the slope of the linear approximation formula is not ≥ 0.001, or if it is determined that the measured values ​​are not twice or more the estimated values, then n = n + 1 is set (S307), and the process returns to S301. In Figures 33, A to F, "○" indicates the measured value, and "×" indicates twice the estimated value.

[0261] [A method that uses two features]

[0262] In this method, the features used are "the proportion of samples where the area of ​​pixels containing both Receiver and Secretion is greater than or equal to a certain value" and "the number of samples where the area of ​​pixels containing only Secretion is greater than or equal to a certain value."

[0263] First, for the first feature (the proportion of samples where the area of ​​pixels containing both Receiver and Secretion is greater than or equal to a certain value), let's assume, for example, that there are 20 samples in the gate shown on the left side of B in Figure 29 and the gate shown on the right side of B in Figure 29. Next, we find the area of ​​each sample that is both Receiver+ and Secretion+ (the target area; see the light gray area in A in Figure 34). Let's assume that there are 18 samples where this area is 17 or less. Then, the feature used is the proportion of samples that are both Receiver+ and Secretion+, which is 18 / 20 = 90%.

[0264] Next, for the second feature (the number of samples where the area of ​​pixels containing only Secretion is greater than or equal to a certain value), let's assume, for example, that there are 30 samples in the gate shown on the left side of Figure 29B and the gate shown on the right side of Figure 29B. Then, let's assume that there are 2 samples where the area of ​​Secretion+ is 200 or greater. In this case, we use this number of 2 samples as the feature. Note that in Figure 34B, the upper panel shows the case where a higher threshold is set, and the lower panel shows the case where a lower threshold is set. In Figure 34B, if the threshold for the area of ​​Secretion+ is set low, the autofluorescence of carriers with a large area (for example, ≥200) will appear as noise, making it impossible to distinguish between Secretion+ samples and Secretion-n samples. Therefore, by using this feature, we avoid this situation.

[0265] Figure 35 is a flowchart showing an example of an embodiment of a method that utilizes two features. The following processing is, in principle, performed in the information processing unit 103. In this embodiment, first, the threshold (for example, six thresholds are set in advance) is set to a high value at the gate shown in A of Figure 35 (S401). Next, the threshold set at the gate shown in A of Figure 35 is applied to all the gates shown in B of Figure 35 (S402). Specifically, all gates from 0 to 8 shown in B of Figure 35 are applied, and the feature quantities are calculated for each obtained image. Next, the feature quantities of the sample group are calculated (S403). Next, the gate with the largest approximate triangle shape where feature quantity 1 (here, the percentage of samples in which the area of ​​pixels in which both Receiver and Secretion exist simultaneously is greater than or equal to a certain value) is X% or more is selected (S404). Here, the value of X can be set to any number, such as 85%.

[0266] Next, it is determined whether the number of features 2 at that gate (here, the number of samples in which the area of ​​pixels containing only Secretion is greater than or equal to a certain value) is V or less (S405). If it is determined to be V or less, the threshold is lowered (S406), and the process returns to S402. On the other hand, if it is determined to be more than V, the gate is determined using the previous threshold and the approximately triangle selected at that time (S407).

[0267] Figures 36 and 37 illustrate an example of the gate determination process. In Figures 36 and 37, the left column shows the graph of feature quantity 1 described above, and the right column shows the graph of feature quantity 2 described above. Figure 36A shows the case where the threshold is set to 360, Figure 36B shows the case where the threshold is set to 350, Figure 36C shows the case where the threshold is set to 340, and Figure 36D shows the case where the threshold is set to 330. Furthermore, Figure 37A shows the case where the threshold is set to 320, Figure 37B shows the case where the threshold is set to 310, Figure 37C shows the case where the threshold is set to 300, and Figure 37D shows the case where the threshold is set to 290. In addition, the numbers 0 to 8 at the bottom of each graph indicate the gate numbers in Figure 35B.

[0268] In the graph of feature 1 in Figure 36 A, the widest gate when X = 85% is gate 4, so gate 4 is selected here. Also, in the graph of feature 2 in Figure 36 A, it can be seen that there are no samples with an area of ​​200 or more at gate 4. As the threshold is lowered further, in Figure 36 C and Figure 36 D, it can be seen that so-called carriers with an area of ​​200 or more enter up to gate 2. Similarly, as the threshold is lowered further, in Figure 37 A and Figure 37 B, carriers enter up to gate 2 or even gate 3.

[0269] Furthermore, as the threshold is lowered further, under the condition that X = 85% or higher, gate #5 is selected in Figure 37 D. However, in Figure 37 D, one carrier with an area of ​​200 or more has entered gate #5, so the specific algorithm in this embodiment is to determine the gate using the previous threshold, the condition "Threshold = 300, Gate = #5" shown in Figure 37 C.

[0270] 100 Particle analyzer 101 Irradiation unit 102 Detection unit 103 Information processing unit 104 Sorting unit 200 Processing unit 1000 Secretion evaluation system

Claims

A position information identification step that identifies the position information of the secretory cell and the antigen-containing particle based on optical information relating to at least one secretory cell and at least one carrier holding an antigen-containing particle, A secretion information acquisition step to acquire information about secretions that bind to unbound molecules possessed by the secretory cells and the antigen-carrying particles, A secretion evaluation step, in which the characteristics of the secretion are evaluated based on the location information and the information regarding the secretion, A method for evaluating secretions, including the following. 　 The secretion evaluation method according to claim 1, further comprising a photodetection step of detecting light from the carrier and obtaining optical information relating to the carrier. 　 The secretion evaluation method according to claim 1, wherein the optical information relating to the carrier is image or fluorescence information. 　 The secretion evaluation method according to claim 1, wherein in the secretion information acquisition step, the difference in signals between unstained secretions and stained secretions, or the signal ratio of unstained secretions and stained secretions, on the secretory cells and the antigen-carrying particles is acquired as information regarding secretions that bind to the unbinding molecules possessed by the secretory cells and the antigen-carrying particles. 　 The secretion evaluation method according to claim 4, further comprising using the evaluation index consisting of subtraction or division of unstained secretions in the secretion information acquisition step. 　 The secretion evaluation method according to claim 1, further comprising a capture molecule placement step of arranging capture molecules that capture the secretion on the carrier. 　 The secretion evaluation method according to claim 6, wherein in the capture molecule placement step, the amount of secretion that is not bound to the binding molecule is detected. 　 The secretion evaluation method according to claim 6, wherein, in the capture molecule placement step, the antigen-containing particles are made to coexist in the carrier in advance, and the amount of secretion that exhibits nonspecific binding to the unbound molecule is measured. 　 The secretion evaluation method according to claim 6, wherein in the capture molecule placement step, the capture molecules are fixed in the carrier in advance, and the amount of secretion that is not bound to the untetherable molecules is measured. 　 In the position information identification step, the carrier further holds at least one antigen-free particle, and the position information of the antigen-free particle is further identified based on the optical information relating to the carrier. The secretion evaluation method according to claim 1, wherein the secretion information acquisition step uses the positional information of the antigen-free particles. 　 The secretion evaluation method according to claim 10, wherein in the secretion information acquisition step, the difference in signals between the non-antigen-containing particles and the antigen-containing particles, or the signal ratio of the non-antigen-containing particles and the antigen-containing particles, is acquired as information concerning secretions that bind to the unbinding molecules possessed by the secretory cells and the antigen-containing particles. 　 In the aforementioned photodetection step, fluorescence from the fluorescently labeled carrier is detected. The secretion evaluation method according to claim 2, wherein in the position information identification step, optical information relating to the carrier is corrected based on the fluorescence intensity of the detected carrier. 　 The secretion evaluation method according to claim 12, wherein in the position information identification step, a fluorescence correction coefficient is calculated from the fluorescence intensity of the detected carrier, and the fluorescence intensity of one or more selected from the group consisting of the secretory cell, the antigen-containing particle, the secretion, and the carrier surface is corrected. 　 The secretion evaluation method according to claim 1, wherein the characteristic is the binding affinity or binding specificity between the binding-tolerant molecule possessed by the antigen-possessing particle and the secretion. 　 The secretion evaluation method according to claim 14, further comprising a sorting step of sorting the secretory cells if the binding specificity or binding affinity is above a certain level. 　 In the location information identification step, the location information of the secretion is further identified, The secretion evaluation method according to claim 1, further comprising a binding state detection step of detecting the state in which the secretion is bound to the unbound molecule possessed by the antigen-possessing particle, based on the relationship between the positional information of the secretory cell, the antigen-possessing particle, and the secretion. 　 The secretion evaluation method according to claim 16, wherein in the binding state detection step, the degree to which the secretion is bound to the unbound molecule possessed by the antigen-possessing particle is quantified based on the fluorescence signals of the secretory cell, the antigen-possessing particle, and the secretion. 　 The method for evaluating secretions according to claim 17, wherein the quantification uses the overlap of each of the fluorescent signals. 　 The secretion evaluation method according to claim 1, wherein in the secretion evaluation step, a signal that changes when the secretion binds to the unbinding molecule possessed by the antigen-possessing particle is detected, and the detected signal is used for the evaluation. 　 A channel through which a liquid containing particles flows, An irradiation unit that irradiates light onto particles flowing through the aforementioned channel, A detection unit for detecting light generated by the aforementioned light irradiation, A particle analyzer equipped with, A positional information identification step that identifies the positional information of the secretory cell and the antigen-containing particle based on optical information relating to at least one secretory cell and a carrier holding at least one antigen-containing particle, A secretion information acquisition step to acquire information about secretions that bind to unbound molecules possessed by the secretory cells and the antigen-carrying particles, A secretion evaluation step, in which the characteristics of the secretion are evaluated based on the location information and the information regarding the secretion, A processing device that performs the following: A secretion evaluation system having the following features.

Images