Information processing apparatus, method of operating the information processing apparatus, and operating program for the information processing apparatus

Abstract

To provide a sensor unit of a spectroscopic analyzer, a measurement system and a measurement method, each enabling reduction of risk of adversely affecting physical property data.SOLUTION: A distal end portion has an inner wall surface and an inner shell portion. The inner wall surface is a wall surface facing an emission surface of a transparent plate. The inner shell portion projects from an emission surface side to an inner wall surface side, and holds the inner wall surface while a space in which a culture supernatant fluid flows between the inner wall surface and the emission surface is provided. The inner shell portion includes an inflow entrance and an outflow port of the culture supernatant fluid, and further includes the inner wall surface that is arranged in both sides in a direction in which the culture supernatant fluid flows.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The technology disclosed herein relates to a sensor unit of a spectroscopic analyzer, a measurement system, and a measurement method. [Background technology]

[0002] The spectroscopic analyzer has a sensor unit. The tip of the sensor unit has an optical system built in for irradiating a measurement light onto a substance to be measured for physical property data and for capturing return light from the substance to be measured. As described in Patent Document 1, the tip may be attached to a flow cell. The flow cell has a flow path through which a fluid containing the substance to be measured flows. The flow cell is a type of container for a fluid containing the substance to be measured. Patent Document 1 describes flow cells having various flow path diameters.

[0003] Patent Document 2 describes a sensor unit having an L-shaped leg at the end of an optical system (transparent element). The leg has an end that faces the emission surface of the measurement light of the optical system at a certain distance. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Special Publication No. 2020-511635 [Patent Document 2] Patent Publication No. 2021-048872 Summary of the Invention [Problem to be solved by the invention]

[0005] As described in Patent Document 1, the flow channel diameter of the flow cell is diverse. Therefore, the distance between the focusing position of the measurement light by the optical system and the wall surface of the flow channel facing the exit surface of the measurement light of the optical system becomes wider or narrower depending on the flow channel diameter. As a result, even if the same measurement target substance is measured, the physical property data will differ depending on the flow channel diameter of the flow cell. Specifically, when the flow channel diameter of the flow cell is relatively large and the distance between the focusing position and the wall surface of the flow channel facing the exit surface is relatively wide, the intensity value of the physical property data will be lower overall than when the flow channel diameter of the flow cell is relatively small and the distance between the focusing position and the wall surface of the flow channel facing the exit surface is relatively narrow.

[0006] In Patent Document 2, a leg is provided with an end portion that faces the emission surface of the measurement light of the optical system at a certain distance. The leg portion in Patent Document 2 has an L-shaped cross section, and is open except for the portion connected to the end portion. For this reason, when the flow cell contains a resin, for example, a new problem occurs in that the measurement light leaking from the open portion of the leg is irradiated onto the resin of the flow cell, and the return light is taken in from the open portion, adversely affecting the physical property data.

[0007] One embodiment of the technique disclosed herein provides a sensor unit of a spectroscopic analyzer, a measurement system, and a measurement method that can reduce the risk of adversely affecting physical property data. [Means for solving the problem]

[0008] The sensor unit of the spectroscopic analysis device disclosed herein is a tip unit incorporating an optical system for irradiating measurement light onto a substance to be measured for physical property data and capturing return light from the substance to be measured, and is a sensor unit of a spectroscopic analysis device having a tip unit to be attached to a container for a fluid containing the substance to be measured, and the tip unit has an opposing wall surface that faces the emission surface of the measurement light of the optical system, and a holding unit that protrudes from the emission surface side to the opposing wall surface side and holds the opposing wall surface with a space provided between the emission surface and the opposing wall surface through which the fluid flows, and the holding unit includes an inlet and outlet for the fluid, and side wall surfaces arranged on both sides in the direction of fluid flow.

[0009] The container is preferably a flow cell having a channel through which a fluid flows.

[0010] The tip is preferably attached to a flow cell containing a resin.

[0011] The tip portion is preferably attached to a plurality of types of flow cells having different flow path diameters.

[0012] It is preferable that at least a portion of the surface of the opposing wall is made of metal.

[0013] The area of ​​the metal on the opposing wall is preferably larger than the area of ​​the opposing wall that is irradiated with the measurement light.

[0014] The surface roughness of the opposing wall surface is preferably 1.6 μm or less.

[0015] The opposing wall surface is preferably flat.

[0016] When the emission surface side is the upper side and the opposing wall side is the lower side, it is preferable that the opposing wall be a curved surface that is convex downward.

[0017] It is preferable that the body portion comprises a resin.

[0018] The inlet and outlet are preferably circular or rectangular.

[0019] The position at which the measurement light is focused by the optical system is preferably between the emission surface and the opposing wall surface.

[0020] The optical system preferably includes a lens having positive refractive power.

[0021] It is preferable that the optical system further includes an optical element having a flat exit surface for the measurement light.

[0022] The thickness of the optical element in the optical axis direction is preferably equal to or greater than half the distance between the point where the exit surface of the lens intersects with the optical axis and the focusing position of the measurement light.

[0023] It is preferable that a metal is disposed on at least a portion of the surface of the side wall.

[0024] It is preferable that the distance between the emission surface and the opposing wall surface is fixed.

[0025] The physical property data is preferably Raman spectral data.

[0026] The turbidity of the fluid is preferably greater than or equal to 250 NTU and less than or equal to 1000 NTU.

[0027] Preferably, the fluid is a cell culture medium.

[0028] A measurement system according to the present disclosure includes a sensor unit of any of the spectroscopic analyzers described above, and a container into which the tip unit is attached.

[0029] When the container is a flow cell and the tip is attached to the flow cell, the flow paths through which the fluid flows are preferably present on the inside and outside of the tip.

[0030] When the container is a flow cell and the tip is attached to the flow cell, it is preferred that a line connecting the centers of the inlet and outlet be parallel to the flow direction.

[0031] The measurement method of the present disclosure measures physical property data using a sensor unit of any of the above-described spectroscopic analyzers. Effect of the Invention

[0032] According to the technology disclosed herein, it is possible to provide a sensor unit of a spectroscopic analysis device, a measurement system, and a measurement method that can reduce the risk of adversely affecting physical property data. [Brief description of the drawings]

[0033] [Figure 1]FIG. 1 is a diagram showing how Raman spectrum data of a measurement target substance in a cell culture solution obtained from a culture tank during culture is measured by a measurement system. [Diagram 2] FIG. 2 is a diagram showing excitation light and Raman scattered light. [Diagram 3] FIG. [Figure 4] FIG. [Diagram 5] FIG. 2 is a cross-sectional view of a flow cell and a sensor unit. [Figure 6] FIG. 2 is a diagram showing the positional relationship between a flow channel of a flow cell and an inlet and an outlet at the tip. [Figure 7] FIG. 13 is a diagram showing an example in which a tip portion is attached to a flow cell having a relatively small flow path diameter. [Figure 8] FIG. 13 is a diagram showing an example in which a tip portion is attached to a flow cell having a medium flow path diameter. [Figure 9] FIG. 13 is a diagram showing an example in which a tip portion is attached to a flow cell having a relatively large flow path diameter. [Figure 10] 4 is an explanatory diagram of the focusing position of excitation light and the thickness of an optical element in the optical axis direction. FIG. [Figure 11] 13 is a diagram showing the relationship between the area of ​​metal on the opposing wall surface and the area of ​​the opposing wall surface irradiated with excitation light. FIG. [Figure 12] FIG. 13 is a diagram showing an opposing wall surface that is curved and convex downward. [Figure 13] FIG. 13 shows a circular inlet and outlet. [Figure 14] FIG. 1 illustrates an information processing apparatus. [Figure 15] FIG. 2 is a block diagram of a computer constituting the information processing device. [Figure 16] 2 is a block diagram showing a processing unit of a CPU of the information processing device. [Figure 17] FIG. 1 is a diagram showing the structure of a data set group. [Figure 18] FIG. 13 is a diagram showing a process in a learning phase of a concentration prediction model. [Figure 19]FIG. 13 is a diagram showing how Raman spectrum data measured using a flow cell with a medium flow path diameter is applied to a concentration prediction model and how a concentration prediction result is output from the concentration prediction model. [Figure 20] FIG. 13 is a diagram showing how Raman spectrum data measured using a flow cell with a relatively large flow path diameter is applied to a concentration prediction model and how a concentration prediction result is output from the concentration prediction model. [Figure 21] FIG. 13 is a diagram showing a Raman spectrum analysis screen. [Figure 22] FIG. 13 is a diagram showing a Raman spectrum analysis screen on which concentration prediction results are displayed. [Diagram 23] 10 is a flowchart showing a processing procedure in a learning phase of a concentration prediction model. [Figure 24] 13 is a flowchart showing a processing procedure of the information processing device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0034] [First embodiment] As an example, as shown in FIG. 1, the measurement system 2 includes a flow cell 10 and a Raman spectrometer 11. The measurement system 2 is incorporated in a cell culture unit 12 in a manufacturing system for an active pharmaceutical ingredient of a biopharmaceutical, for example. The cell culture unit 12 has a culture tank 13 and a cell removal filter 14. A cell culture solution 15 is stored in the culture tank 13. The flow cell 10 is an example of a "container" according to the technology of the present disclosure. The Raman spectrometer 11 is an example of a "spectroscopic analysis device" according to the technology of the present disclosure. The cell removal filter 14 may be an alternating tangential flow filtration (ATF) filter.

[0035] Antibody-producing cells 16 are seeded in the culture tank 13, and the antibody-producing cells 16 are cultured in a cell culture solution 15. The antibody-producing cells 16 are cells established by incorporating an antibody gene into a host cell such as a Chinese Hamster Ovary cell (CHO cell). The antibody-producing cells 16 produce immunoglobulin, i.e., an antibody 17, during the culture process. Therefore, not only the antibody-producing cells 16 but also the antibody 17 are present in the cell culture solution 15. The antibody 17 is, for example, a monoclonal antibody, and serves as an active ingredient in a biopharmaceutical. The antibody 17 is an example of a "substance to be measured" according to the technology of the present disclosure.

[0036] A first delivery line 18 is connected to the culture tank 13. A cell removal filter 14 is disposed in the first delivery line 18. The cell removal filter 14 captures antibody-producing cells 16 in the cell culture fluid 15 with a filter membrane (not shown) by, for example, a tangential flow filtration (TFF) method, and removes the antibody-producing cells 16 from the cell culture fluid 15. The cell removal filter 14 also allows antibodies 17 to pass through. Therefore, the cell culture fluid 15 containing mainly antibodies 17 flows downstream of the cell removal filter 14 in the first delivery line 18. The cell culture fluid 15 from which the antibody-producing cells 16 have been removed by the cell removal filter 14 in this manner is called a culture supernatant. Hereinafter, the cell culture fluid 15 from which the antibody-producing cells 16 have been removed by the cell removal filter 14 is referred to as a culture supernatant 15A. The culture supernatant 15A is an example of a "fluid" according to the technology of the present disclosure.

[0037] The turbidity of the culture supernatant 15A is 250 NTU (Nephelometric Turbidity Unit) or more and 1000 NTU or less. NTU is a unit of turbidity of a liquid based on a formazin standard solution. In addition to the antibody 17, the culture supernatant 15A also contains cell-derived proteins, cell-derived DNA (Deoxyribonucleic Acid), aggregates of the antibody 17, viruses, etc. These cell-derived proteins, cell-derived DNA, aggregates of the antibody 17, and viruses are also examples of the "substance to be measured" according to the technology of the present disclosure.

[0038] The flow cell 10 is connected to a first outlet path 18 downstream of the cell removal filter 14. As indicated by arrow FD, culture supernatant liquid 15A flows from the first outlet path 18 at a preset flow rate through the flow cell 10. A delivery pump (not shown) is provided downstream of the cell removal filter 14 on the first outlet path 18 (between the cell removal filter 14 and the flow cell 10). The delivery pump delivers the culture supernatant liquid 15A toward the flow cell 10 at a flow rate of 200 cc / min or more, for example 300 cc / min.

[0039] A second outlet path 19 is also connected to the flow cell 10. The culture supernatant 15A that flows into the flow cell 10 from the first outlet path 18 flows out to the second outlet path 19. The second outlet path 19 is connected to a purification section that purifies the antibody 17 from the culture supernatant 15A using, for example, a chromatography device, and sends the culture supernatant 15A from the flow cell 10 to the purification section.

[0040] As an example, as shown in FIG. 2, the Raman spectrometer 11 is an instrument for evaluating a substance M by utilizing the characteristics of the Raman scattered light RSL. When the substance M is irradiated with the excitation light EL, the excitation light EL interacts with the substance M to generate Raman scattered light RSL having a wavelength different from that of the excitation light EL. The wavelength difference between the excitation light EL and the Raman scattered light RSL corresponds to the energy of the molecular vibration of the substance M. For this reason, it is possible to obtain Raman scattered light RSL having different wave numbers between substances M having different molecular structures. The excitation light EL is an example of the "measurement light" according to the technology of the present disclosure. The Raman scattered light RSL is an example of the "return light" according to the technology of the present disclosure. Of the Stokes line and the anti-Stokes line, it is preferable to use the Stokes line as the Raman scattered light RSL.

[0041] Returning to FIG. 1, the Raman spectrometer 11 is composed of a sensor unit 25 and an analyzer 26. The sensor unit 25 is connected to the flow cell 10. The sensor unit 25 emits excitation light EL from its tip. The excitation light EL is irradiated onto the culture supernatant fluid 15A flowing inside the flow cell 10. Raman scattered light RSL is generated by interaction between this excitation light EL and the antibodies 17, etc. in the culture supernatant fluid 15A. The sensor unit 25 receives the Raman scattered light RSL and outputs the received Raman scattered light RSL to the analyzer 26.

[0042] The analyzer 26 resolves the Raman scattered light RSL into wavenumbers and derives the intensity value of the Raman scattered light RSL for each wavenumber, thereby generating Raman spectrum data 27. The Raman spectrum data 27 is data in which the intensity value of the Raman scattered light RSL for each wavenumber is registered. In FIG. 1, the Raman spectrum data 27 is -1 ~1800cm -1 The intensity values ​​of the Raman scattered light RSL in the range from -1 The data is derived at intervals. The graph shown below the Raman spectrum data 27 is obtained by plotting the intensity values ​​of the Raman spectrum data 27 for each wave number and connecting them with a line. The Raman spectrum data 27 is an example of the "physical property data" according to the technology of the present disclosure.

[0043] In this manner, the measurement system 2 causes the culture supernatant 15A obtained from the culture tank 13 in which the antibody-producing cells 16 are being cultured to flow through the flow cell 10. Then, the culture supernatant 15A flowing through the flow cell 10 is irradiated with excitation light EL through the sensor unit 25, thereby measuring Raman spectrum data 27 of the antibodies 17, etc. in the culture supernatant 15A.

[0044] As an example, as shown in FIG. 3, the flow cell 10 is a cylindrical member having a linear flow path 30 with a circular cross section at the center. The flow cell 10 contains resin. The resin content in the flow cell 10 is 95% or more, for example, 99%. Alternatively, the resin content is 100%, and the flow cell 10 is entirely made of resin. In other words, the flow cell 10 is made of resin. The resin is, for example, a polyolefin resin. In this case, the flow cell 10 may be single-use. The flow cell 10 may be made of metal.

[0045] A cylindrical boss-shaped first connection part 31 and a second connection part 32 are provided at the center of both end faces of the flow cell 10. The first connection part 31 has an inlet 33 of the flow path 30, and the second connection part 32 has an outlet 34 of the flow path 30. The direction parallel to the flow path 30 from the inlet 33 to the outlet 34 is the flow direction FD of the culture supernatant 15A. The direction FD is an example of a "flow direction" according to the technology of the present disclosure.

[0046] The first connection part 31 and the second connection part 32 are parallel threads or tapered threads. A sterile connector 35 provided at one end of the first delivery path 18 is liquid-tightly attached to the first connection part 31. A sterile connector 36 provided at one end of the second delivery path 19 is liquid-tightly attached to the second connection part 32. Ferrules may be used to connect the first connection part 31 and the second connection part 32 to the sterile connectors 35 and 36.

[0047] A mounting portion 37 is provided at the center of the peripheral plate of the flow cell 10. The mounting portion 37 is a cylindrical hole for detachably mounting the sensor portion 25 to the flow cell 10, and has a thread 38 on its inner wall surface. The mounting portion 37 is provided up to the flow path 30. A part of the sensor portion 25 is housed in the mounting portion 37.

[0048] The sensor unit 25 includes a main body 39 and a tip 40. The main body 39 is a cylindrical member having a linear light passage 41 with a circular cross section at the center. The excitation light EL and the Raman scattered light RSL pass through the light passage 41. The excitation light EL passes through the light passage 41 from the main body 39 to the tip 40. Conversely, the Raman scattered light RSL passes through the light passage 41 from the tip 40 to the main body 39 and, ultimately, to the analyzer 26. The main body 39 contains resin, similar to the flow cell 10. The resin content in the main body 39 is 95% or more, for example, 99%. Alternatively, the resin content is 100%, and the main body 39 is entirely made of resin. In other words, the main body 39 is made of resin. The resin is, for example, a polyolefin resin, similar to the flow cell 10.

[0049] As shown in FIG. 4 as an example, the main body 39 has a large diameter portion 45, a medium diameter portion 46, and a small diameter portion 47 in order from the base end side. The large diameter portion 45 is the portion of the main body 39 with the largest diameter. The medium diameter portion 46 is a portion with a smaller diameter than the large diameter portion 45 and a larger diameter than the small diameter portion 47. A thread 48 is cut in the medium diameter portion 46. The thread 48 is screwed into the thread 38 of the mounting portion 37. In other words, the main body 39, and in turn the sensor portion 25, are detachably attached to the mounting portion 37 by the medium diameter portion 46. Therefore, the medium diameter portion 46 and the tip portion 40 and the small diameter portion 47, which are the portions of the sensor portion 25 beyond the medium diameter portion 46, are accommodated in the flow cell 10. The small diameter portion 47 is the portion of the main body 39 with the smallest diameter. A thread 49 is cut in the small diameter portion 47.

[0050] The tip portion 40 has an optical system 55, an inner shell portion 56, and an outer shell portion 57. The optical system 55 is between the small diameter portion 47 of the main body portion 39 and the inner shell portion 56, and is held within the outer shell portion 57. The optical system 55 includes a hemispherical lens 58 and a transparent plate 59. The hemispherical lens 58 is literally a hemispherical lens, and is made of, for example, quartz glass. The hemispherical lens 58 has an exit surface 60 for the excitation light EL. The exit surface 60 is flat. The hemispherical lens 58 is an example of a "lens having positive refractive power" according to the technology of the present disclosure.

[0051] The transparent plate 59 is a circular plate having an exit surface 61 and an entrance surface 62 for the excitation light EL that are parallel to each other, and is made of, for example, quartz glass. The exit surface 60 of the hemispherical lens 58 and the entrance surface 62 of the transparent plate 59 have the same curvature (0 in this case). The exit surface 60 and the entrance surface 62 may be fixed and joined by an adhesive or the like, or may simply be held in a state where the surfaces are joined together without using an adhesive or the like. The transparent plate 59 is an example of an "optical element" according to the technology of the present disclosure. The exit surface 61 of the transparent plate 59 is an example of an "exit surface of the measurement light of the optical system" according to the technology of the present disclosure.

[0052] When the side of the main body 39 (the side of the exit surface 61 of the transparent plate 59) is considered as the upper side and the side of the tip portion 40 (the side of the inner wall surface 63A of the bottom plate 63 of the inner shell portion 56) is considered as the lower side (see also FIG. 5), both the inner shell portion 56 and the outer shell portion 57 have a cylindrical container shape with an open upper side and a closed lower side by the flat bottom plates 63 and 64. The inner shell portion 56 has a peripheral plate 65, and the outer shell portion 57 has a peripheral plate 66. The peripheral plate 65 is erected upward from the bottom plate 63, and forms a cylindrical space 67 together with the bottom plate 63. Similarly, the peripheral plate 66 is erected upward from the bottom plate 64, and forms a cylindrical space 68 together with the bottom plate 64.

[0053] The height of peripheral plate 65 is lower than that of peripheral plate 66. In other words, the height of inner shell portion 56 is lower than that of outer shell portion 57. Inner shell portion 56 has a size that is slightly smaller than outer shell portion 57 overall, and is housed tightly within space 68 of outer shell portion 57 (see also FIG. 5).

[0054] The inner shell portion 56 is made of metal, for example, Hastelloy. On the other hand, the outer shell portion 57 contains resin, similar to the flow cell 10, etc. The resin content in the outer shell portion 57 is 95% or more, for example, 99%. Alternatively, the resin content is 100%, and the entire outer shell portion 57 is made of resin. In other words, the outer shell portion 57 is made of resin. The resin is, for example, a polyolefin-based resin, similar to the flow cell 10, etc.

[0055] On the upper side of the inner wall surface 66A of the peripheral plate 66 of the outer shell portion 57, a screw 69 is cut. The screw 69 is screwed with the screw 49 of the small-diameter portion 47. Thereby, the outer shell portion 57, and thus the tip portion 40, is integrated with the main body portion 39.

[0056] Rectangular notches 70A and 71A are formed at 180° symmetric positions on the upper side of the peripheral plate 65. Also, rectangular holes 70B and 71B are formed at 180° symmetric positions at the center of the peripheral plate 66. The holes 70B and 71B are more precisely square-shaped. The notches 70A and 71A and the holes 70B and 71B have the same size. The notch 70A corresponds to the hole 70B, and the notch 71A corresponds to the hole 71B. The notch 70A and the hole 70B function as the inlet 70 for the culture supernatant 17. Also, the notch 71A and the hole 71B function as the outlet 71 for the culture supernatant 17.

[0057] The inner wall surface 63A of the bottom plate 63 of the inner shell portion 56 is planar. The inner wall surface 63A is located on the side toward which the excitation light EL travels (downstream side in the irradiation direction of the excitation light EL), and the excitation light EL is irradiated onto the inner wall surface 63A (see FIGS. 10 and 11). It faces the emission surface 61 of the transparent plate 59. That is, the inner wall surface 63A is an example of the "opposing wall surface" according to the technology of the present disclosure. When the optical system 55 and the inner shell portion 56 are housed in the space 68 of the outer shell portion 57 and the screw 69 of the outer shell portion 57 is screwed with the screw 49 of the small-diameter portion 47 so that the tip portion 40 is integrated with the main body portion 39, the distance between the emission surface 61 and the inner wall surface 63A is fixed.

[0058] The surface roughness (arithmetic mean roughness) Ra of the inner wall surface 63A is greater than 0 and 6.4 μm or less (0 < Ra ≤ 6.4 μm), preferably 1.6 μm or less (0 < Ra ≤ 1.6 μm). Although not shown, the surface roughness Ra of the inner wall surface 65A is also greater than 0 and 6.4 μm or less, preferably 1.6 μm or less. The inner wall surfaces 63A and 65A are subjected to a smoothing process such as polishing to make the surface roughness Ra a target value. Note that the surface roughness Ra is a value measured in accordance with JIS B 0601-2001 defined by the Japanese Industrial Standards.

[0059] When viewed from above, the inner wall surface 65A of the peripheral plate 65 of the inner shell part 56 protrudes from the exit surface 61 side of the transparent plate 59 to the inner wall surface 63A side. The inner wall surface 65A holds the inner wall surface 63A with a space 67 through which the culture supernatant fluid 17 flows between the exit surface 61 and the inner wall surface 65A. The inner wall surface 65A is disposed on both sides of the direction FD and blocks both sides of the direction FD. That is, the inner wall surface 65A is an example of a "side wall surface" according to the technology of the present disclosure, and the inner shell part 56 is an example of a "holding part" according to the technology of the present disclosure. As can be seen from the inner shell part 56, the functions of the "opposing wall surface" and the "holding part" according to the technology of the present disclosure may be performed by one member. The reference symbol 64A indicates the inner wall surface of the bottom plate 64 of the outer shell part 57.

[0060] As an example, as shown in FIG. 5, the optical system 55 is held so as to be sandwiched between the tip of the small diameter portion 47 of the main body portion 39 and the edge of the peripheral plate 65 of the inner shell portion 56. The height of the emission surface 61 of the transparent plate 59 and the edges of the upper ends of the inlet 70 and the outlet 71 are the same. A circular groove 80 is formed in the part of the mounting portion 37 that contacts the medium diameter portion 46 of the main body portion 39. An O-ring 81 is fitted in the groove 80. The O-ring 81 is made of elastic rubber, and is crushed between the medium diameter portion 46 and the bottom plate of the groove 80 by fastening and fixing the main body portion 39 to the mounting portion 37 with the screws 38 and 48. The O-ring 81 is crushed between the medium diameter portion 46 and the bottom plate of the groove 80 to prevent the culture supernatant 15A flowing through the flow path 30 from leaking out of the mounting portion 37. Although not shown, O-rings for preventing leakage of the culture supernatant 15A are also disposed between the small diameter portion 47 and the hemispherical lens 58, and between the inner shell portion 56 and the transparent plate 59.

[0061] 5, when the tip portion 40 is attached to the flow cell 10, a line LIO connecting the center CI of the inlet 70 and the center CO of the outlet 71 is parallel to the direction FD. In other words, the line LIO is parallel to the center line of the flow channel 30. Here, "parallel" refers to parallelism in the sense of including, in addition to perfect parallelism, an error that is generally acceptable in the technical field to which the technology of the present disclosure belongs and that does not go against the spirit of the technology of the present disclosure.

[0062] 6, when the tip 40 is attached to the flow cell 10, the center CI of the inlet 70, the center CO of the outlet 71, and the center CP of the flow channel 30 coincide. Here, "coincidence" refers to a perfect coincidence as well as a coincidence including an error that is generally acceptable in the technical field to which the technology of the present disclosure belongs and that does not contradict the spirit of the technology of the present disclosure. The error here is preferably ±10%, more preferably ±5%.

[0063] As can be seen from FIG. 6, when the tip 40 is attached to the flow cell 10, the flow path through which the culture supernatant 15A flows exists inside and outside the tip 40. In other words, the culture supernatant 15A flows inside and outside the tip 40. The flow path inside the tip 40 is defined by the inlet 70 and the outlet 71, the exit surface 61 of the transparent plate 59, the inner wall surface 63A of the bottom plate 63 of the inner shell 56, and the inner wall surface 65A of the peripheral plate 65 of the inner shell 56. The flow path outside the tip 40 is defined by the flow path 30 of the flow cell 10, the outer wall surface 64B of the bottom plate 64 of the outer shell 57, and the outer wall surface 66B of the peripheral plate 66 of the outer shell 57. The area of ​​the inlet 70 and the outlet 71 is preferably 1 / 2 or less, more preferably 1 / 4 or less, and even more preferably 1 / 8 or less of the area of ​​the peripheral plate 65 of the inner shell 56.

[0064] As an example, as shown in Figs. 7 to 9, the tip 40 is attached to a plurality of types of flow cells 10 having different flow path diameters. Specifically, as shown in Fig. 7, the tip 40 is attached to a flow cell 10S having a flow path 30S having a flow path diameter φS. As shown in Fig. 8, the tip 40 is attached to a flow cell 10M having a flow path 30M having a flow path diameter φM. As shown in Fig. 9, the tip 40 is attached to a flow cell 10L having a flow path 30L having a flow path diameter φL. The flow path diameters φS, φM, and φL are in a relationship of φS<φM<φL. That is, the flow cell 10S has a relatively small flow path diameter φ, the flow cell 10M has a medium flow path diameter φ, and the flow cell 10L has a relatively large flow path diameter φ. The flow rate of the culture supernatant 15A is always the same regardless of the type of flow cell 10.

[0065] Here, when the tip portion 40 is attached to any of the flow cells 10S, 10M, and 10L, the line LIO connecting the center CI of the inlet 70 and the center CO of the outlet 71 is parallel to the direction FD. When the tip portion 40 is attached to any of the flow cells 10S, 10M, and 10L, the center CI of the inlet 70, the center CO of the outlet 71, and the center CP of the flow path 30 are the same. Here, three types of flow cells 10S, 10M, and 10L are illustrated, but the number of types of flow cells 10 to which the tip portion 40 is attached may be two or four or more.

[0066] 10, the excitation light EL emitted from the exit surface 61 of the transparent plate 59 is focused at a focusing position FP located within the inner shell portion 56. The focal length is determined by the diameter and refractive index of the hemispherical lens 58, and the focusing position FP is determined by the focal length. In this case, the focusing position FP is point-like. The focusing position FP is located between the exit surface 61 of the transparent plate 59 and the inner wall surface 63A of the bottom plate 63 of the inner shell portion 56.

[0067] When the thickness of the transparent plate 59 in the direction of the optical axis OA is Th and the distance between the first point P1 intersecting the optical axis OA on the exit surface 60 of the hemispherical lens 58 and the focusing position FP is d, the thickness Th is more than half the distance d and is less than or equal to the distance d (d / 2≦Th≦d). More preferably, the thickness Th is the same as the distance d, that is, Th=d. When Th=d, the focusing position FP coincides with the second point P2 intersecting the optical axis OA on the exit surface 61 of the transparent plate 59. Note that the thickness Th and the distance d are "the same" in the sense that they are completely the same, as well as including an error that is generally acceptable in the technical field to which the technology of the present disclosure belongs and that does not go against the spirit of the technology of the present disclosure. The error here is preferably ±10%, more preferably ±5%.

[0068] As an example, as shown in Fig. 11, the excitation light EL that has passed through the focusing position FP spreads in a cone shape from the focusing position FP toward the inner wall surface 63A, and is finally irradiated onto the inner wall surface 63A. Reference numeral 90 denotes an irradiation area of ​​the excitation light EL on the inner wall surface 63A. The irradiation area 90 is a circle centered on the optical axis OA. The diameter of this irradiation area 90 is smaller than the diameter of the inner wall surface 63A. In other words, the area of ​​the metal on the inner wall surface 63A is larger than the irradiation area of ​​the excitation light EL on the inner wall surface 63A (area of ​​the metal on the inner wall surface 63A>irradiation area of ​​the excitation light EL on the inner wall surface 63A).

[0069] Next, the operation of the above-mentioned configuration will be described. The measurement system 2 consisting of the flow cell 10 and the Raman spectrometer 11 is incorporated into the cell culture section 12. The flow cell 10 has a first connection section 31 connected to the first outlet path 18 and a second connection section 32 connected to the second outlet path 19. The sensor section 25 (tip section 40) of the Raman spectrometer 11 attached to the flow cell 10 via the attachment section 37. A culture supernatant 15A obtained from a culture tank 13 in which antibody-producing cells 16 are being cultured is caused to flow through the flow path 30 of the flow cell 10. The culture supernatant 15A flows from an inlet 70 into the tip section 40 of the sensor section 25 and flows out of the tip section 40 from an outlet 71. In the tip section 40, the culture supernatant 15A is irradiated with excitation light EL that has passed through a light path 41 and an optical system 55. The excitation light EL is focused at a focusing position FP by the optical system 55.

[0070] Raman scattered light RSL is generated by the interaction between the excitation light EL and the antibodies 17, etc. in the culture supernatant 15A. The Raman scattered light RSL is taken in by the optical system 55, passes through the optical path 41, and is output from the sensor unit 25 to the analyzer 26. The Raman scattered light RSL is converted into Raman spectrum data 27 by the analyzer 26.

[0071] The tip 40 has an inner wall surface 63A and an inner shell portion 56. The inner wall surface 63A is a wall surface facing the exit surface 61 of the transparent plate 59. The inner shell portion 56 protrudes from the exit surface 61 side to the inner wall surface 63A side, and holds the inner wall surface 63A with a space 67 through which the culture supernatant liquid 15A flows between the exit surface 61 and the inner shell portion 56. The inner shell portion 56 includes an inlet 70 and an outlet 71 for the culture supernatant liquid 15A, and inner wall surfaces 65A arranged on both sides in the direction FD. This inner wall surface 65A can reduce the risk of adversely affecting the Raman spectrum data 27 compared to the sensor portion described in Patent Document 2, in which one side in the fluid flow direction is open. For example, the excitation light EL is irradiated onto a resin flow cell 10, and the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin of the flow cell 10 is captured, thereby reducing the risk of adversely affecting the Raman spectrum data 27. Even if the flow cell 10 is not made of resin, it is possible to reduce the risk that the Raman spectrum data 27 will be adversely affected by Raman scattered light from the flow cell 10 due to leakage of the excitation light EL.

[0072] The intensity of the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin of the flow cell 10 is relatively high. For this reason, in the case of the sensor unit described in Patent Document 2, the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin of the flow cell 10 drowns out the Raman scattered light RSL generated by the interaction between the excitation light EL and the antibody 17, etc. in the culture supernatant 15A, which should be measured. In other words, the S / N ratio of the Raman spectrum data 27 due to the Raman scattered light RSL generated by the interaction between the excitation light EL and the antibody 17, etc. in the culture supernatant 15A is significantly reduced.

[0073] However, in the technology disclosed herein, since there are inner wall surfaces 65A disposed on both sides in the direction FD, there is little risk that the Raman scattered light RSL generated by the interaction between the excitation light EL and the antibodies 17, etc. in the culture supernatant 15A, which should be measured, will be drowned out by the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin of the flow cell 10. In other words, the S / N ratio of the Raman spectrum data 27 due to the Raman scattered light RSL generated by the interaction between the excitation light EL and the antibodies 17, etc. in the culture supernatant 15A can be maintained at a higher level.

[0074] Furthermore, the presence of the inner wall surfaces 65A disposed on both sides of the direction FD can stabilize the flow of the culture supernatant 15A near the focusing position FP of the excitation light EL in the inner shell portion 56, compared to the sensor portion described in Patent Document 2. This reduces bias in components in the culture supernatant 15A near the focusing position FP, and can improve the measurement stability of the Raman spectrum data 27.

[0075] If a branch flow path of a preset flow path diameter is provided in the first delivery path 18 and the flow cell 10 is connected to the branch flow path, Raman spectrum data 27 can be measured using the flow cell 10 of the same diameter regardless of the flow path diameter of the first delivery path 18. However, it is necessary to provide a branch flow path each time, which is time-consuming. In contrast, the technology of the present disclosure does not require such time-consuming work.

[0076] Furthermore, as will be described in detail later in the second embodiment, according to the sensor unit 25, a concentration prediction model 126 (see FIG. 16) for predicting the concentration of the antibody 17 contained in the culture supernatant 15A based on the Raman spectrum data 27 can be generated only from the Raman spectrum data 27 measured using one type of flow cell 10, and the generated concentration prediction model 126 can be shared by multiple types of flow cells 10. This can save the effort of preparing a concentration prediction model 126 for each of the multiple types of flow cells 10.

[0077] 1 and the like, the container to which the tip portion 40 is attached is a flow cell 10 having a flow path 30 through which the culture supernatant 15A flows. Therefore, the Raman spectrum data 27 of the culture supernatant 15A can be easily measured without modifying the culture tank 13, such as providing an attachment portion for the tip portion 40.

[0078] 3, the tip 40 is attached to a flow cell 10 containing a resin. This significantly reduces the risk that the excitation light EL is irradiated onto the resin of the flow cell 10 and the resulting Raman scattered light RSL is captured and adversely affects the Raman spectrum data 27.

[0079] 7 to 9, the tip portion 40 is attached to a plurality of types of flow cells 10S, 10M, and 10L having different flow path diameters. As described above, the tip portion 40 has an inner wall surface 63A facing the exit surface 61 of the transparent plate 59, and the distance between the exit surface 61 of the transparent plate 59 and the inner wall surface 63A does not change. Therefore, even if the same substance to be measured is measured, the Raman spectrum data 27 does not differ between the plurality of types of flow cells 10S, 10M, and 10L having different flow path diameters, and the tip portion 40 can be used without any problems with flow cells 10 having a wide variety of flow path diameters.

[0080] 4, the inner shell portion 56 is made of metal, and the inner wall surfaces 63A and 65A are also formed of metal. This makes it possible to further reduce the risk that the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin when the excitation light EL is irradiated onto the resin will be captured and adversely affect the Raman spectrum data 27.

[0081] 11, the area of ​​the metal on the inner wall surface 63A is larger than the area of ​​the inner wall surface 63A irradiated with the excitation light EL. For this reason, it is possible to more reliably reduce the risk that the Raman scattered light RSL generated by the interaction between the excitation light EL and the resin when the resin is irradiated with the excitation light EL is captured and adversely affects the Raman spectrum data 27.

[0082] 4, the surface roughness Ra of the inner wall surface 63A is 1.6 μm or less. Therefore, the excitation light EL and the Raman scattered light RSL are reflected by the inner wall surface 63A, and this enables the intensity value of the Raman spectrum data 27 to be increased.

[0083] 4, the inner wall surface 63A is a flat surface, which makes it possible to easily manufacture the inner wall surface 63A, and therefore the inner shell portion 56.

[0084] 3, the main body 39 contains resin, which allows the main body 39 to be manufactured at low cost and can be easily disposed of.

[0085] 4 and other drawings, the inlet 70 and the outlet 71 are rectangular. This allows the culture supernatant 15A to flow stress-free toward the focusing position FP of the excitation light EL in the inner shell 56. This reduces bias in components in the culture supernatant 15A near the focusing position FP, thereby improving the measurement stability of the Raman spectrum data 27.

[0086] 10, the focusing position FP of the excitation light EL by the optical system 55 is between the exit surface 61 and the inner wall surface 63A of the transparent plate 59. Therefore, the Raman spectrum data 27 of the culture supernatant fluid 15A flowing inside the inner shell portion 56 can be reliably measured.

[0087] 4 and other drawings, the optical system 55 includes a hemispherical lens 58 having a positive refractive power. Therefore, the focusing position FP of the excitation light EL can be set to a position relatively close to the exit surface 61 of the transparent plate 59, and the possibility that the excitation light EL will be attenuated by the culture supernatant fluid 15A can be reduced.

[0088] 4 and other drawings, the optical system 55 further includes a transparent plate 59 having a flat exit surface 61 for the excitation light EL. This can further reduce the risk that the excitation light EL will be attenuated by the culture supernatant fluid 15A.

[0089] 10, the thickness Th of the transparent plate 59 in the direction of the optical axis OA is equal to or more than half the distance d between a first point P1 intersecting the optical axis OA on the exit surface 60 of the hemispherical lens 58 and a focusing position FP of the excitation light EL, but is equal to or less than the distance d. If the thickness Th is equal to or more than half the distance d, the effect of reducing the risk that the excitation light EL will be attenuated by the culture supernatant fluid 15A can be more effectively achieved. If the thickness Th is equal to or less than the distance d, the focusing position FP can be always outside the transparent plate 59.

[0090] The distance between the exit surface 61 of the transparent plate 59 and the inner wall surface 63A is fixed. Therefore, even if the same substance to be measured is measured, the Raman spectrum data 27 will not differ between the multiple types of flow cells 10S, 10M, and 10L having different flow path diameters, and the flow cells 10 having a wide variety of flow path diameters can be used without any problems.

[0091] Raman scattered light RSL easily reflects information derived from the functional groups of amino acids in proteins. Therefore, by using Raman spectrum data 27 as the physical property data as in this example, it is possible to obtain physical property data that accurately reflects the physical properties, such as the concentration of antibody 17, which is a protein.

[0092] As shown in Fig. 1, the turbidity of the culture supernatant 15A is 250 NTU or more and 1000 NTU or less. In this case, the attenuation of the excitation light EL by the culture supernatant 15A becomes greater. Therefore, by using a hemispherical lens 58 with a positive refractive power or a transparent plate 59 with a flat exit surface 61 for the excitation light EL, it is possible to greatly reduce the risk of the excitation light EL being attenuated by the culture supernatant 15A.

[0093] Biopharmaceuticals containing the cell product antibody 17 are called antibody drugs and are widely used to treat chronic diseases such as cancer, diabetes, and rheumatoid arthritis, as well as rare diseases such as hemophilia and Crohn's disease. Therefore, according to this example in which the culture supernatant 15A, which is the source of the antibody drug obtained from the culture tank 13 in which the antibody-producing cells 16 are cultured, is used as the fluid, it is possible to promote the development of antibody drugs that are widely used to treat various diseases.

[0094] The fluid used is culture supernatant 15A obtained from culture tank 13 during culture. Therefore, Raman spectrum data 27 can be measured while continuing to culture antibody-producing cells 16 in culture tank 13. Furthermore, culture supernatant 15A is from which antibody-producing cells 16 have been removed. Therefore, Raman spectrum data 27 of the cell products, such as antibodies 17, can be measured with high accuracy.

[0095] 6 and other drawings, when the tip portion 40 is attached to the flow cell 10, the flow path through which the culture supernatant liquid 15A flows exists on the inside and outside of the tip portion 40. Therefore, the Raman spectrum data 27 of the culture supernatant liquid 15A can be measured within the tip portion 40 (inner shell portion 56). Furthermore, the tip portion 40 does not impede the flow of the culture supernatant liquid 15A in the flow cell 10.

[0096] 5, when the tip 40 is attached to the flow cell 10, a line LIO connecting the centers CI and CO of the inlet 70 and outlet 71 is parallel to the direction FD. This allows the culture supernatant 15A to flow without stress toward the focusing position FP of the excitation light EL in the inner shell 56. This reduces bias in the components of the culture supernatant 15A in the inner shell 56, thereby improving the measurement stability of the Raman spectrum data 27.

[0097] (Variation 1) One example may be a tip portion 100 as shown in Fig. 12. The tip portion 100 has an inner shell portion 101. A bottom plate 102 of the inner shell portion 101, and further an inner wall surface 102A of the bottom plate 102, is a downwardly convex curved surface that follows the shape of the flow channel 30 of the flow cell 10. The inner wall surface 102A is an example of an "opposing wall surface" according to the technology of the present disclosure.

[0098] In this way, by making the inner wall surface 102A a downwardly convex curved surface, the inner wall surface 102A plays a role of a reflecting surface that directs the Raman scattered light RSL toward the optical system 55. This makes it possible to further increase the S / N ratio of the Raman spectrum data 27. The downwardly convex curved surface may be in the shape of a parabolic antenna.

[0099] (Variation 2) As an example, the tip portion 105 may be as shown in Fig. 13. The tip portion 105 has a circular inlet 106 and an outlet 107. In this way, the shapes of the inlet and the outlet are not limited to the rectangular shapes of the inlet 70 and the outlet 71, but may be circular shapes of the inlet 106 and the outlet 107.

[0100] The inner shell portion 56 is not limited to the cylindrical container shape shown in the example. It may be a square or hexagonal container shape, etc. Therefore, the bottom plate 63 is not limited to the circular shape shown in the example, but may be a rectangular or hexagonal shape, etc. Furthermore, the peripheral plate 66 is not limited to the curved surface shown in the example, but may be a flat surface.

[0101] [Second embodiment] In the second embodiment, the state of the measurement target substance such as the antibody 17 contained in the culture supernatant 15A is predicted based on the Raman spectrum data 27.

[0102] 14, the Raman spectrometer 11 is connected to an information processing device 110 via a computer network such as a LAN (Local Area Network) so as to be able to communicate with each other. The Raman spectrometer 11 transmits Raman spectrum data 27 to the information processing device 110. The information processing device 110 is, for example, a desktop personal computer, a notebook personal computer, or a tablet terminal.

[0103] 15, as an example, a computer constituting an information processing device 110 includes a storage 115, a memory 116, a CPU (Central Processing Unit) 117, a communication unit 118, a display 119, and an input device 120. These are interconnected via a bus line 121.

[0104] The storage 115 is a hard disk drive built into a computer constituting the information processing device 110 or connected via a cable or a network. Alternatively, the storage 115 is a disk array in which a plurality of hard disk drives are connected in series. The storage 115 stores control programs such as an operating system, various application programs, and various data associated with these programs. Note that a solid state drive may be used instead of the hard disk drive.

[0105] The memory 116 is a work memory for the CPU 117 to execute processing. The CPU 117 loads a program stored in the storage 115 into the memory 116 and executes processing according to the program. In this way, the CPU 117 comprehensively controls each part of the computer. The memory 116 may be built into the CPU 117.

[0106] The communication unit 118 is a network interface that controls the transmission of various information via a LAN or the like. The display 119 displays various screens. The various screens are provided with an operation function using a GUI (Graphical User Interface). The computer constituting the information processing device 110 accepts input of operation instructions from the input device 120 via the various screens. The input device 120 is a keyboard, a mouse, a touch panel, a microphone for voice input, etc.

[0107] 16 as an example, an operating program 125 is stored in the storage 115 of the information processing device 110. The operating program 125 is an application program for causing a computer to function as the information processing device 110. In addition to the operating program 125, a concentration prediction model 126 is also stored in the storage 115. The concentration prediction model 126 is a machine learning model configured by, for example, a neural network. Note that the model is not limited to a neural network, and may be a decision tree, a random forest, a naive Bayes, a gradient boosting decision tree, or the like.

[0108] When the operating program 125 is started, the CPU 117 of the computer constituting the information processing device 110 functions as an acquisition section 130, a RW control section 131, a prediction section 132, and a display control section 133 in cooperation with the memory 116 and the like.

[0109] The acquiring unit 130 acquires the Raman spectrum data 27 from the Raman spectrometer 11. The acquiring unit 130 outputs the Raman spectrum data 27 to the RW control unit 131.

[0110] The RW control unit 131 controls the storage of various data in the storage 115 and the reading of various data stored in the storage 115. The RW control unit 131 stores the Raman spectrum data 27 from the acquisition unit 130 in the storage 115. The RW control unit 131 also reads out the Raman spectrum data 27 and the concentration prediction model 126 from the storage 115, and outputs the read out Raman spectrum data 27 and the concentration prediction model 126 to the prediction unit 132. The RW control unit 131 also outputs the Raman spectrum data 27 to the display control unit 133.

[0111] The prediction unit 132 applies the Raman spectrum data 27 to the concentration prediction model 126, and causes the concentration prediction model 126 to output a concentration prediction result 135. The concentration prediction result 135 is a result of predicting the concentration of the antibody 17 in the culture supernatant 15A. The prediction unit 132 outputs the concentration prediction result 135 to the display control unit 133.

[0112] The display control unit 133 controls the display of various screens on the display 119. For example, the display control unit 133 controls the display of a Raman spectrum analysis screen 150 (see FIG. 21 and the like) on the display 119.

[0113] 17, the data set 140 is a set of a learning intensity value 141 and a correct concentration 142. A data set group 140G, which is a collection of the data sets 140, is prepared for the learning phase of the concentration prediction model 126.

[0114] The learning intensity value 141 is a copy of the intensity value of the Raman spectrum data 27S measured using a flow cell 10S having a flow channel 30S with a flow channel diameter φS shown in Fig. 7. The correct concentration 142 is calculated based on the amount of antibody 17 in the culture supernatant 15A from which the Raman spectrum data 27S was measured, and the volume of the flow channel 30S. The amount of antibody 17 is measured using, for example, a high performance liquid chromatography device.

[0115] 18, in the learning phase of the concentration prediction model 126, a learning intensity value 141 from a data set 140 is input to the concentration prediction model 126, and a learning concentration prediction result 135L is output from the concentration prediction model 126. Next, a loss calculation is performed for the concentration prediction model 126 using a loss function based on a comparison result between the learning concentration prediction result 135L and a correct concentration 142. Then, update settings of the coefficients of the concentration prediction model 126 are performed according to the result of the loss calculation, and the concentration prediction model 126 is updated according to the update settings.

[0116] In the learning phase of the concentration prediction model 126, the above-mentioned series of processes, including input of the learning intensity value 141 to the concentration prediction model 126, output of the learning concentration prediction result 135L from the concentration prediction model 126, loss calculation, update setting, and update of the concentration prediction model 126, are repeatedly performed while changing the data set 140. The repetition of the above-mentioned series of processes is terminated when the prediction accuracy of the learning concentration prediction result 135L for the correct concentration 142 reaches a predetermined set level. The concentration prediction model 126 whose prediction accuracy has thus reached the set level is stored in the storage 115 and used by the prediction unit 132. Note that learning may be terminated when the above-mentioned series of processes have been repeated a set number of times, regardless of the prediction accuracy of the learning concentration prediction result 135L for the correct concentration 142.

[0117] The concentration prediction model 126 may be trained in the information processing device 110, or in a device other than the information processing device 110. In addition, even after the concentration prediction model 126 is stored in the storage 115, the training of the concentration prediction model 126 may be continued.

[0118] The prediction unit 132 applies the Raman spectrum data 27 measured using a plurality of types of flow cells 10 different from the flow cell 10S, which have a flow path 30 with a flow path diameter φ different from the flow path diameter φS, to the concentration prediction model 126. Then, the concentration prediction model 126 outputs a concentration prediction result 135. More specifically, as shown in FIG. 19 as an example, the prediction unit 132 inputs the intensity value of the Raman spectrum data 27M measured using the flow cell 10M having the flow path 30M with the flow path diameter φM shown in FIG. 8 to the concentration prediction model 126. Then, the concentration prediction model 126 outputs a concentration prediction result 135. Also, as shown in FIG. 20 as an example, the prediction unit 132 inputs the intensity value of the Raman spectrum data 27L measured using the flow cell 10L having the flow path 30L with the flow path diameter φL shown in FIG. 9 to the concentration prediction model 126. Then, the concentration prediction model 126 outputs a concentration prediction result 135. Although not shown in the figure, the prediction unit 132 inputs the intensity value of the Raman spectrum data 27S measured using a flow cell 10S having a flow path 30S with a flow path diameter φS into the concentration prediction model 126, and outputs the concentration prediction result 135 from the concentration prediction model 126.

[0119] 21 on the display 119. On the Raman spectrum analysis screen 150, a graph of the Raman spectrum data 27 is displayed.

[0120] A concentration prediction button 151 is provided at the bottom of the Raman spectrum analysis screen 150. When the concentration prediction button 151 is pressed, a concentration prediction instruction is accepted by the CPU 117 of the information processing device 110. Upon receiving the concentration prediction instruction, the CPU 117 causes the prediction unit 132 to perform the processes shown in Fig. 19 and Fig. 20 and output the concentration prediction result 135 from the concentration prediction model 126.

[0121] When the concentration prediction result 135 is input from the prediction unit 132, the display control unit 133 transitions the display of the Raman spectrum analysis screen 150 to an example as shown in Fig. 22. In Fig. 22, the concentration prediction result 135 is displayed on the Raman spectrum analysis screen 150 together with a graph of the Raman spectrum data 27.

[0122] Next, the operation of the second embodiment will be described with reference to the flow charts shown in FIGS. 23 and 24 as an example.

[0123] First, the concentration prediction model 126 is trained as shown in FIG. 18 using the data set 140 shown in FIG. 17. That is, the learning intensity value 141, which is a copy of the intensity value of the Raman spectrum data 27S measured using a flow cell 10S having a flow path 30S with a flow path diameter φS, is input to the concentration prediction model 126, and the learning concentration prediction result 135L is output from the concentration prediction model 126 (step ST100 in FIG. 23). Next, the concentration prediction model 126 is updated based on the comparison result between the learning concentration prediction result 135L and the correct concentration 142 (step ST110). The processes of steps ST100 and ST110 are repeatedly performed while the data set 140 is being changed (step ST130) until the prediction accuracy of the learning concentration prediction result 135L with respect to the correct concentration 142 reaches a predetermined set level (NO in step ST120). When the prediction accuracy of the learning concentration prediction result 135L for the correct concentration 142 reaches the set level (YES in step ST120), the learning of the concentration prediction model 126 is terminated. The concentration prediction model 126 for which the learning has been completed is stored in the storage 115 of the information processing device 110.

[0124] As shown in FIG. 16, the CPU 117 of the information processing device 110 functions as an acquisition unit 130, a RW control unit 131, a prediction unit 132, and a display control unit 133 by starting the operating program 125.

[0125] A concentration prediction model 126 is stored in the storage 115 of the information processing device 110. The concentration prediction model 126 is read out from the storage 115 by the RW control unit 131 and output to the prediction unit 132.

[0126] In the information processing device 110, the Raman spectrum data 27 from the Raman spectrometer 11 is acquired by the acquisition unit 130 (step ST200 in FIG. 24). The Raman spectrum data 27 is stored in the storage 115 by the RW control unit 131 (step ST210).

[0127] The Raman spectrum data 27 is read from the storage 115 by the RW control unit 131 (step ST220), and output to the prediction unit 132 and the display control unit 133. Then, as shown in Fig. 21, the display control unit 133 displays a Raman spectrum analysis screen 150 on the display 119 (step ST230).

[0128] The user of the information processing device 110 presses the concentration prediction button 151 in order to cause the concentration prediction model 126 to predict the concentration of the antibody 17 in the culture supernatant 15A from which the Raman spectrum data 27, the graph of which is displayed on the Raman spectrum analysis screen 150, has been measured. This causes the concentration prediction instruction to be accepted by the CPU 117 (step ST240).

[0129] 19 and 20, the prediction unit 132 inputs the intensity value of the Raman spectrum data 27 to the concentration prediction model 126, which in turn outputs a concentration prediction result 135 (step ST250). The concentration prediction result 135 is output from the prediction unit 132 to the display control unit 133, and is displayed on the Raman spectrum analysis screen 150 by the display control unit 133, as shown in FIG. 22 (step ST260).

[0130] The user makes various decisions with reference to the concentration prediction result 135 displayed on the Raman spectrum analysis screen 150. For example, consider a case where a condition-finding experiment is being conducted on the culture conditions of the antibody-producing cells 16 using a small-scale facility. In this case, if the concentration prediction result 135 is worse than the target value, the user decides to stop the current experiment and move on to an experiment using new conditions. Also consider a case where the condition-finding experiment is completed and mass production is being conducted using a large-scale facility. In this case, if the concentration prediction result 135 is worse than the target value, the user decides to stop mass production and perform maintenance on the culture tank 13.

[0131] As described above, in the second embodiment, the CPU 117 of the information processing device 110 uses a concentration prediction model 126 that predicts the concentration of the antibody 17 contained in the culture supernatant 15A. The concentration prediction model 126 is generated using only a data set 140 that is composed of learning intensity values ​​141 obtained by copying intensity values ​​of Raman spectrum data 27S of the antibody 17, etc., contained in the culture supernatant 15A in the flow cell 10S, and a correct concentration 142 of the antibody 17.

[0132] The acquisition unit 130 acquires Raman spectrum data 27M and 27L of the antibody 17, etc., contained in the culture supernatant 15A in a flow cell 10 different from the flow cell 10S, for example, in the flow cells 10M and 10L. The prediction unit 132 applies the Raman spectrum data 27M and 27L to the concentration prediction model 126, and causes the concentration prediction model 126 to output a concentration prediction result 135 of the antibody 17.

[0133] Concentration prediction model 126 generated only from Raman spectrum data 27S measured using flow cell 10S can be used to predict the concentration of antibody 17 based on Raman spectrum data 27M and 27L measured using flow cells 10M and 10L that are different from flow cell 10S. This eliminates the need to prepare concentration prediction models 126 for multiple types of flow cells 10, such as for flow cell 10S, flow cell 10M, and flow cell 10L.

[0134] All that is required as the learning intensity values ​​141 of the data set 140 are the intensity values ​​of the Raman spectrum data 27S of the antibody 17 etc. contained in the culture supernatant 15A in the flow cell 10S. This makes it possible to easily prepare the data set 140.

[0135] The Raman spectrum data 27S, 27M, and 27L are measured by the sensor unit 25 having the tip portion 40 shown in the first embodiment. Therefore, even if the same measurement target substance is measured, the Raman spectrum data 27S, 27M, and 27L do not differ from each other, and there is no need to correct the Raman spectrum data 27S, 27M, and 27L to eliminate the differences between them.

[0136] The concentration prediction model 126 is not limited to a machine learning model. It may be a model generated by multivariate analysis or statistical analysis. Examples of multivariate analysis or statistical analysis include multiple regression, principal component regression, partial least squares regression, logistic regression, Lasso regression, ridge regression, support vector regression, and Gaussian process regression. In such a model generated by multivariate analysis or statistical analysis, determining the coefficient of a regression equation based on at least two datasets 140 corresponds to "generating a state prediction model" using only datasets" in Appendix 25, etc., which will be described later.

[0137] The Raman spectrum data 27 used to generate the data set 140 is not limited to the illustrated Raman spectrum data 27S. The intensity values ​​of the Raman spectrum data 27M or the intensity values ​​of the Raman spectrum data 27L may be used as the learning intensity values ​​141 of the data set 140.

[0138] The flow rate of the culture supernatant 15A may differ depending on the type of the flow cell 10. In this case, however, it is preferable to apply the Raman spectrum data 27 to the concentration prediction model 126 after correcting the difference in the Raman spectrum data 27 caused by the difference in the flow rate of the culture supernatant 15A. When the Raman spectrum data 27 used to generate the data set 140 is, for example, the Raman spectrum data 27S, a correction is performed to make the Raman spectrum data 27M and 27L equivalent to the Raman spectrum data 27S. For the correction, a conversion formula that makes the Raman spectrum data 27M and 27L equivalent to the Raman spectrum data 27 or a machine learning model may be used.

[0139] Although the concentration of antibody 17 in culture supernatant 15A is predicted, the present invention is not limited to this. The concentration of aggregates in culture supernatant 15A may also be predicted. Moreover, instead of or in addition to the concentration, density or the like may also be predicted.

[0140] The container to which the tip portion 40 is attached is not limited to the flow cell 10. The container to which the tip portion 40 is attached may be, for example, a culture tank 13. In this case, a mounting portion is provided in the culture tank 13, and the tip portion 40 is attached to the mounting portion. When the container is the culture tank 13, the flow direction is the direction in which the cell culture fluid 15 flows, which is generated by the rotation of the stirring blades in the culture tank 13.

[0141] Although the inner wall surface 63A of the bottom plate 63 of the inner shell part 56 is used as the opposing wall surface, this is not limited thereto. The tip part 40 may be formed only by the outer shell part 57 without using the inner shell part 56, and the inner wall surface 64A of the bottom plate 64 of the outer shell part 57 may be used as the opposing wall surface. In this case, the outer shell part 57 is made of metal, like the inner shell part 56. Alternatively, the outer shell part 57 is made of resin, and a metal film such as aluminum, copper, or gold is formed on the entire surface or a part of the inner wall surface 64A by plating. Furthermore, a metal film is also formed on the entire surface or a part of the inner wall surface 66A of the peripheral plate 66 of the outer shell part 57 by plating. In this way, it is sufficient that a metal is disposed on at least a part of the surface of the opposing wall surface and at least a part of the surface of the side wall surface. However, in order to fully exert the effect of reducing the risk that the S / N ratio of the Raman spectrum data 27 will be reduced due to the resin, it is preferable that a metal is disposed on the entire surface of the opposing wall surface and the entire surface of the side wall surface.

[0142] The functions of the "opposing wall surface" and the "retaining portion" according to the technology of the present disclosure are performed by inner shell portion 56, which is a single member, but are not limited to this. Bottom plate 63 having inner wall surface 63A, which is the opposing wall surface, and other parts of inner shell portion 56 that perform the function of the retaining portion may originally be separate bodies and may be integrated by bonding, welding, or the like.

[0143] The lens having a positive refractive power is not limited to the exemplified hemispherical lens 58. It may be a spherical ball lens, a plano-convex lens, a biconvex lens, a cylindrical lens, or the like. In addition, the optical element is not limited to the exemplified disc-shaped transparent plate 59 having parallel exit surface 61 and entrance surface 62. It may be a transparent plate having an entrance surface shaped to match the shape of the exit surface of a ball lens, a plano-convex lens, a biconvex lens, or the like, and a flat exit surface.

[0144] Instead of forming an optical system by cementing a lens having a positive refractive power and an optical element, the lens having a positive refractive power and the optical element may be integrally formed as one lens. It should be noted that the optical element may not be required, and the optical system may be constructed only with lenses having positive refractive power.

[0145] A mechanism may be provided that changes the distance between the exit surface of the optical system 55 for the excitation light EL (exit surface 61 of the transparent plate 59) and the opposing wall surface (inner wall surface 63A) in a stepwise manner, for example. In this case, the distance between the exit surface of the optical system 55 for the excitation light EL and the opposing wall surface is set according to the flow cell 10 having the smallest flow path diameter φ among the flow cells 10 expected to be used. In other words, the distance between the exit surface of the optical system 55 for the excitation light EL and the opposing wall surface does not need to be fixed.

[0146] The first connecting part 31 and the second connecting part 32 may be disposed below the flow cell 10, and the flow path 30 may be U-shaped. The shape of the flow cell 10 is not limited to a cylindrical shape, and may be a square tube shape. The cross-sectional shape of the flow path 30 is not limited to a circular shape, and may be an elliptical shape or a rectangular shape. The flow cell 10 may be formed of a composite material such as carbon fiber reinforced resin.

[0147] The substance for which the Raman spectrum data 27 is measured is not limited to the antibody 17, etc. Other than the antibody 17, a protein, a peptide, a nucleic acid (DNA, RNA (Ribonucleic Acid)), a lipid, a virus, a virus subunit, a virus-like particle, etc. may be used.

[0148] The cell product is not limited to antibody 17, etc. It may be a cytokine (interferon, interleukin, etc.), a hormone (insulin, glucagon, follicle-stimulating hormone, erythropoietin, etc.), a growth factor (IGF (insulin-like growth factor)-1, bFGF (basic fibroblast growth factor), etc.), a blood coagulation factor (factor 7, factor 8, factor 9, etc.), an enzyme (lysosomal enzyme, DNA (deoxyribonucleic acid) degrading enzyme, etc.), an Fc (fragment crystallizable) fusion protein, a receptor, albumin, or a protein vaccine. In addition, the antibody 17 may also be a bispecific antibody, an antibody-drug conjugate, a low molecular weight antibody, a glycosylated antibody, etc.

[0149] The physical property data is not limited to the Raman spectrum data 27. It may be infrared absorption spectrum data, near-infrared absorption spectrum data, nuclear magnetic resonance spectrum data, ultraviolet-visible absorption spectroscopy (UV-Vis) spectrum data, or fluorescence spectrum data.

[0150] The fluid is not limited to the culture supernatant 15A. It may be the cell culture fluid 15 before the cells are removed by the cell removal filter 14. It may be a cell culture fluid (so-called culture medium) that does not contain cell products before being supplied to the culture tank 13. It may be a purified liquid obtained by purifying the culture supernatant 15A with a chromatography device in the purification section. The fluid is not limited to the cell culture fluid 15, and may be, for example, river water collected to check water pollution. It may be a raw material (e.g., polystyryl lithium and an aqueous methanol solution, etc.) and / or a product when continuously producing a product such as a monomer or polymer (e.g., polystyrene, etc.) by flow synthesis. In addition, the fluid is not limited to a liquid and may be a gas.

[0151] When the tip 40 is attached to the flow cell 10, the center CI of the inlet 70, the center CO of the outlet 71, and the center CP of the flow path 30 are aligned, but this is not limited thereto. When the flow path diameter φ is relatively large, the flow cell 10 and the sensor unit 25, etc. may be designed so that the center CI of the inlet 70 and the center CO of the outlet 71 are positioned above the center CP of the flow path 30. In this way, the amount of protrusion of the tip 40, etc. into the flow path 30 can be reduced, and the resistance of the tip 40, etc. to the flow of the culture supernatant fluid 15A can be reduced. Alternatively, the amount of protrusion of the tip 40, etc. into the flow path 30 may be the same regardless of the type of flow cell 10.

[0152] From the above description, the technology described in the following supplementary paragraphs can be understood.

[0153] [Additional note 1] A sensor unit of a spectroscopic analysis device including a tip portion that is attached to a container of a fluid containing the target substance, the tip portion having a built-in optical system for irradiating a measurement light onto a target substance for measuring physical property data and capturing return light from the target substance, the sensor unit comprising: The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having The sensor part of a spectroscopic analyzer. [Additional note 2] 2. The sensor unit of the spectroscopic analysis device according to claim 1, wherein the container is a flow cell having a flow path through which the fluid flows. [Additional note 3] 3. The sensor portion of the spectroscopic analysis device according to claim 2, wherein the tip portion is attached to the flow cell containing a resin. [Additional note 4] The sensor unit of the spectroscopic analyzer according to claim 2 or 3, wherein the tip portion is attached to a plurality of types of the flow cells having different flow path diameters. [Additional note 5] The sensor unit of the spectroscopic analyzer according to any one of claims 1 to 4, wherein a metal is disposed on at least a part of a surface of the opposing wall. [Additional note 6] 6. The sensor unit of the spectroscopic analysis device according to claim 5, wherein an area of ​​the metal on the opposing wall is larger than an area of ​​the opposing wall irradiated with the measurement light. [Additional note 7] 7. The sensor unit of the spectroscopic analyzer according to any one of claims 1 to 6, wherein the opposing wall surface has a surface roughness of 1.6 μm or less. [Additional note 8] The sensor unit of the spectroscopic analyzer according to any one of supplementary items 1 to 7, wherein the opposing wall surface is a flat surface. [Additional note 9] The sensor unit of the spectroscopic analysis device according to any one of supplementary items 1 to 7, wherein when the exit surface side is the upper side and the opposing wall side is the lower side, the opposing wall surface is a curved surface that is convex toward the lower side. [Additional Note 10] 10. A sensor unit of the spectroscopic analyzer according to any one of claims 1 to 9, comprising a body portion containing resin. [Additional Note 11] The sensor unit of the spectroscopic analyzer according to any one of supplementary items 1 to 10, wherein the inlet and the outlet are circular or rectangular. [Additional Note 12] 12. The sensor unit of the spectroscopic analysis device according to claim 1, wherein a focusing position of the measurement light by the optical system is between the exit surface and the opposing wall surface. [Additional Note 13] 13. The sensor unit of the spectroscopic analyzer according to claim 1, wherein the optical system includes a lens having a positive refractive power. [Additional Note 14] 14. The sensor unit of the spectroscopic analysis device according to claim 13, wherein the optical system further includes an optical element having a flat exit surface for the measurement light. [Additional Note 15] A sensor unit of a spectroscopic analysis device described in appended claim 14, wherein a thickness of the optical element in the optical axis direction is equal to or greater than half the distance between a point where the optical axis intersects with the exit surface of the lens and a focusing position of the measurement light, and is equal to or less than said distance. [Additional Note 16] 16. The sensor unit of the spectroscopic analyzer according to any one of claims 1 to 15, wherein a metal is disposed on at least a part of a surface of the side wall. [Additional Note 17] 17. The sensor unit of the spectroscopic analysis device according to any one of supplementary items 1 to 16, wherein a distance between the emission surface and the opposing wall surface is fixed. [Additional Note 18] 18. The sensor unit of the spectroscopic analysis device according to any one of claims 1 to 17, wherein the physical property data is Raman spectrum data. [Additional Note 19] 19. The sensor unit of the spectroscopic analyzer according to any one of claims 1 to 18, wherein the turbidity of the fluid is 250 NTU or more and 1000 NTU or less. [Additional Note 20] 20. The sensor unit of the spectroscopic analysis device according to any one of claims 1 to 19, wherein the fluid is a cell culture medium. [Additional Note 21] A sensor unit of a spectroscopic analyzer according to any one of claims 1 to 20; The container to which the tip portion is attached; and A measurement system comprising: [Additional note 22] 22. The measurement system according to claim 21, wherein the container is the flow cell, and when the tip is attached to the flow cell, a flow path through which the fluid flows is present on the inside and outside of the tip. [Additional Note 23] 23. The measurement system according to claim 21, wherein when the container is the flow cell and the tip is attached to the flow cell, a line connecting the centers of the inlet and the outlet is parallel to the flow direction. [Additional note 24] A measurement method for measuring the physical property data by using a sensor unit of the spectroscopic analysis device according to any one of claims 1 to 20.

[0154] Moreover, from the description of the second embodiment above, the technology described in the following supplementary claims can be understood.

[0155] [Additional note 25] A processor is provided. The processor, A state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance, acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; Applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measured substance from the state prediction model; The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion that is attached to the first container and the second container, the tip portion including an optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, Information processing device. [Additional note 26] the first container is a first flow cell having a first flow path diameter and a first flow path through which the fluid flows; 26. The information processing device according to claim 25, wherein the second container is a second flow cell having a second flow path through which the fluid flows, the second flow path having a second flow path diameter different from the first flow path diameter. [Additional note 27] a state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance; acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; and applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measurement target substance from the state prediction model; Including, The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion that is attached to the first container and the second container, the tip portion including an optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, A method for operating an information processing device. [Additional note 28] a state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance; acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; and applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measured substance from the state prediction model; An operating program for an information processing device for causing a computer to execute a process including the steps of: The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion that is attached to the first container and the second container, the tip portion including an optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, An operating program for an information processing device.

[0156] Here, the CPU 117 is an example of the above-mentioned "processor." The concentration prediction model 126 is an example of the above-mentioned "state prediction model." The concentration is an example of the above-mentioned "state." The flow cell 10S is an example of the above-mentioned "first container" and "first flow cell." The Raman spectrum data 27S and the learning intensity value 141 are an example of the above-mentioned "first physical property data." The correct concentration 142 is an example of the above-mentioned "correct data."

[0157] The flow cells 10M and 10L are an example of the above-mentioned "second container." The Raman spectrum data 27M and 27L are an example of the above-mentioned "second physical property data." The concentration prediction result 135 is an example of the above-mentioned "prediction result."

[0158] In the second embodiment, for example, the following various processors can be used as the hardware structure of the processing unit that executes various processes, such as the acquisition unit 130, the RW control unit 131, the prediction unit 132, the display control unit 133, and the correction unit 165. As described above, the various processors include the CPU 117, which is a general-purpose processor that executes software (operation program 125) and functions as various processing units, as well as a programmable logic device (PLD), which is a processor whose circuit configuration can be changed after manufacture, such as an FPGA (Field Programmable Gate Array), a dedicated electric circuit, which is a processor having a circuit configuration designed specifically for executing specific processes, such as an ASIC (Application Specific Integrated Circuit), etc.

[0159] A single processing unit may be composed of one of these various processors, or may be composed of a combination of two or more processors of the same or different types (e.g., a combination of multiple FPGAs and / or a combination of a CPU and an FPGA). Also, multiple processing units may be composed of a single processor.

[0160] As an example of configuring multiple processing units with one processor, first, there is a form in which one processor is configured with a combination of one or more CPUs and software, as typified by computers such as client and server, and this processor functions as multiple processing units. Second, there is a form in which a processor is used that realizes the functions of the entire system including multiple processing units with one IC (Integrated Circuit) chip, as typified by System On Chip (SoC), etc. In this way, the various processing units are configured using one or more of the above various processors as a hardware structure.

[0161] Furthermore, more specifically, the hardware structure of these various processors can be an electric circuit that combines circuit elements such as semiconductor elements.

[0162] The technology of the present disclosure can be appropriately combined with the above-mentioned various embodiments and / or various modified examples. In addition, it is needless to say that the technology is not limited to the above-mentioned embodiments, and various configurations can be adopted without departing from the gist of the technology.

[0163] The above description and illustrations are detailed descriptions of the parts related to the technology of the present disclosure, and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, function, action, and effect is an example of the configuration, function, action, and effect of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replaced with respect to the description and illustrations shown above, within the scope of the gist of the technology of the present disclosure. In addition, in order to avoid confusion and to facilitate understanding of the parts related to the technology of the present disclosure, the description and illustrations shown above omit explanations of technical common sense that do not require explanation in order to enable the implementation of the technology of the present disclosure.

[0164] In this specification, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. In addition, in this specification, the same idea as "A and / or B" is also applied when three or more things are expressed by connecting them with "and / or."

[0165] All publications, patent applications, and standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, and standard was specifically and individually indicated to be incorporated by reference. [Explanation of symbols]

[0166] 2. Measurement System 10, 10L, 10M, 10S flow cells 11 Raman spectrometer 12 Cell culture department 13 Culture tank 14 Cell removal filter 15 Cell culture medium 15A Culture supernatant 16 Antibody-producing cells 17 Antibodies 18 1st delivery route 19 2nd delivery route 25 Sensor section 26 Analyzer 27, 27L, 27M, 27S Raman spectral data 30, 30L, 30M, 30S flow path 31 First connection part 32 Second connection part 33 Inlet 34 Outlet 35, 36 Sterile connector 37 Mounting part 38, 48, 49, 69 screws 39 Main body 40, 100, 105 tip 41 Light passage 45 Large diameter section 46 Medium diameter part 47 Small diameter section 55 Optical system 56, 101 Inner shell 57 Outer shell 58 Hemispherical Lens 59 Transparent plate 60 Excitation light exit surface of hemispherical lens 61 Transparent plate excitation light exit surface 62 Transparent plate, excitation light incident surface 63, 102 Bottom plate of inner shell 63A, 102A Inner wall surface of the bottom plate of the inner shell 64 Bottom plate of outer shell 64A Inner wall surface of the bottom plate of the outer shell 64B Outer wall surface of bottom plate of outer shell 65 Inner shell peripheral plate 65A Inner wall surface of the peripheral plate of the inner shell 66 Outer shell peripheral plate 66A Inner wall surface of outer shell peripheral plate 66B Outer wall surface of the outer shell peripheral plate 67 Inner shell space 68 Shell Space 70, 106 Inlet 70A, 71A cutout 70B, 71B hole 71, 107 Outlet 80 grooves 81 O-ring 90 irradiation area 110 Information processing device 115 Storage 116 Memory 117 CPU 118 Communications Department 119 Display 120 Input Devices 121 Bus Line 125 Operating Program 126 Concentration Prediction Model 130 Acquisition Department 131 Read / write control unit (RW control unit) 132 Prediction Department 133 Display control section 135 Concentration prediction results 135L Learning concentration prediction results 140 Datasets 140G Data Sets 141 Learning strength values 142 Correct Answer Concentration 150 Raman Spectral Analysis Screen 151 Concentration prediction button Center of CI inlet CO outlet center CP Center of flow path d Distance between the first point and the focal point EL excitation light FD Flow direction of culture supernatant FP focusing position LIO Line connecting the centers of the inlet and outlet M substance OA optical axis P1 1st point P2 2nd point RSL Raman scattering light ST100, ST110, ST120, ST130, ST200, ST210, ST220, ST230, ST240, ST250, ST260 Step Th: Thickness of the transparent plate along the optical axis φ, φL, φM, φS Flow cell channel diameter

Claims

1. A processor is provided. The processor, A state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance, acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; Applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measured substance from the state prediction model; The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion having a built-in optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, the tip portion being attached to the first container and the second container, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, Information processing device.

2. the first container is a first flow cell having a first flow path having a first flow path diameter and through which the fluid flows; The information processing device according to claim 1 , wherein the second container is a second flow cell having a second flow path through which the fluid flows, the second flow path having a second flow path diameter different from the first flow path diameter.

3. The information processing apparatus according to claim 1 , wherein the first physical property data and the second physical property data are Raman spectrum data.

4. The information processing device according to claim 1 , wherein the fluid is a cell culture medium.

5. a state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance; acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; and applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measurement target substance from the state prediction model; Including, The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion having a built-in optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, the tip portion being attached to the first container and the second container, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, A method for operating an information processing device.

6. a state prediction model for predicting a state of a target substance contained in a fluid, the state prediction model being generated using only a data set consisting of first physical property data of the target substance contained in the fluid in a first container and ground truth data of the state of the target substance; acquiring second physical property data of the measurement target substance contained in the fluid in a second container different from the first container; and applying the second physical property data to the state prediction model, and outputting a prediction result of the state of the measurement target substance from the state prediction model; An operating program for an information processing device for causing a computer to execute a process including the steps of: The first physical property data and the second physical property data are a sensor unit of a spectroscopic analyzer including a tip portion having a built-in optical system for irradiating the measurement target substance with measurement light and capturing return light from the measurement target substance, the tip portion being attached to the first container and the second container, The tip portion is an opposing wall surface facing an exit surface of the measurement light of the optical system; a holding portion that protrudes from the exit surface side to the opposing wall surface side and holds the opposing wall surface in a state in which a space through which the fluid flows is provided between the exit surface and the opposing wall surface, the holding portion including an inlet and an outlet for the fluid, and side wall surfaces arranged on both sides in a flow direction of the fluid; having Measured by a sensor unit of the spectroscopic analyzer, An operating program for an information processing device.

Images