WO2020162601A1 - Cell type estimation method, cell type estimation device, cell production method, cell production device, monitoring method, monitoring device, learned model production method, and learned model production device - Google Patents

Abstract

Provided is a cell type estimation method whereby a cell type can be estimated in a state where a target cell is alive.　The cell type estimation method according to the present invention comprises: an acquisition step for acquiring a signal light from a test cell, said signal light being acquired using a coherent anti-Stokes Raman scattering (CARS) microscope; and an estimation step for estimating the cell type of the test cell on the basis of the signal light.

Description

Cell type estimation method, cell type estimation apparatus, cell manufacturing method, cell manufacturing apparatus, observation method, observation apparatus, learned model manufacturing method, and learned model manufacturing apparatus

The present invention relates to a cell type estimating method, a cell type estimating apparatus, a cell manufacturing method, a cell manufacturing apparatus, an observing method, an observing apparatus, a learned model manufacturing method, and a learned model manufacturing apparatus.

It has been attempted to evaluate the differentiation potential of pluripotent cells such as iPS cells based on the outer shape of cells (Non-patent documents 1 and 2, Patent document 1). However, there is a problem that it is difficult to evaluate the cell type only by the outer shape of the cell.

Japanese Patent No. 6448129

As another method of estimating cell types, there is a method of collecting target cells and estimating based on the gene expression pattern of the collected cells. However, since it is necessary to lyse and use the cells in order to confirm the gene expression of the target cells, there is a problem that it is difficult to estimate the cell type in the live state of the target cells.

Therefore, an object of the present invention is to provide a method for estimating a cell type that allows the cell type to be estimated in a living state of the target cell.

In order to achieve the above object, a method of estimating a cell type of the present invention (hereinafter, also referred to as “estimation method”) is a signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for a test cell. An acquisition process for acquiring
An estimation step of estimating the cell type of the test cell based on the signal light.

A cell type estimation device of the present invention (hereinafter, also referred to as “estimation device”) is an acquisition unit that acquires signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for a test cell,
And an estimation unit that estimates the cell type of the test cell based on the signal light.

The method for producing cells of the present invention comprises an observation step of observing test cells,
An estimation step of estimating the cell type of the test cell,
And a selection step of selecting test cells of a predetermined cell type,
The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
The estimation step is performed by the method of estimating the cell type of the present invention.

The cell manufacturing apparatus of the present invention includes an observation unit capable of observing a test cell,
An estimation unit for estimating the cell type of the test cell,
Including a selection unit for selecting test cells of a predetermined cell type,
The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
The estimation unit includes the cell type estimation device of the present invention.

The observation method using coherent anti-Stokes Raman scattering (CARS) of the present invention (hereinafter referred to as “observation method”) includes an observation step of observing a test cell,
An estimation step of estimating the cell type of the test cell,
Including a re-observation step of observing the test cells of a predetermined cell type again,
The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
The estimation step is performed by the method of estimating the cell type of the present invention.

An observation apparatus using coherent anti-Stokes Raman scattering (CARS) of the present invention (hereinafter referred to as “observation apparatus”) is an observation unit capable of observing a test cell,
An estimation unit for estimating the cell type of the test cell,
A control unit capable of controlling the observation unit,
The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
The estimation unit includes a cell type estimation device of the present invention,
The control unit causes the observation unit to re-observe the cells estimated to be a predetermined cell type among the test cells.

The method of manufacturing a trained model used for estimating the cell type of the present invention (hereinafter, also referred to as a “learning method”) acquires signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell. Acquisition process to
A learning step of generating a learned model that outputs the estimation result of the cell type of the test cell from the signal light using the pair of the signal light and the cell type of the test cell as teacher data.

A trained model manufacturing device (hereinafter, also referred to as a “learning device”) used for estimating a cell type of the present invention acquires signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell. Acquisition means to
And a learning unit that generates a learned model that outputs an estimation result of the cell type of the test cell from the signal light, using the pair of the signal light and the cell type of the test cell as teacher data.

According to the present invention, it is possible to estimate the cell type while the target cell is alive.

FIG. 1 is a block diagram illustrating an example of the configuration of the estimation device according to the first embodiment. FIG. 2 is a block diagram showing an example of the hardware configuration of the estimation device according to the first embodiment. FIG. 3 is a flowchart showing an example of the estimation method of the first embodiment. FIG. 4 is a block diagram showing an example of the configuration of the manufacturing apparatus according to the second embodiment. FIG. 5 is a flowchart showing an example of the manufacturing method of the second embodiment. FIG. 6 is a block diagram showing an example of the configuration of the observation device of the third embodiment. FIG. 7 is a flowchart showing an example of the observation method of the third embodiment. FIG. 8 is a block diagram showing an example of the configuration of the learning device according to the fourth exemplary embodiment. FIG. 9 is a flowchart showing an example of the learning method of the fourth embodiment. FIG. 10 is a schematic diagram showing the configuration of the optical system of the CARS microscope used in Example 1. FIG. 11 is a photograph showing a CH ₂ stretch image and an SHG image in Example 1.

In the present invention, the “cell” means, for example, an isolated cell, a cell mass (spheroid) composed of cells, a tissue, or an organ. The cells may be, for example, cultured cells or cells isolated from a living body. The cell mass, tissue or organ may be, for example, a cell mass, cell sheet, tissue or organ prepared from the cells, or may be a cell mass, tissue or organ isolated from a living body.

In the present invention, “selection of cells” may be used to mean any of recovery of desired cells, maintenance of desired cells, and removal of cells other than desired cells.

Hereinafter, the present invention will be described in detail with reference to the drawings. However, the present invention is not limited to the following description. It should be noted that, in the following FIGS. 1 to 11, the same parts are denoted by the same reference numerals and the description thereof may be omitted. Further, in the drawings, for convenience of description, the structure of each part may be simplified as appropriate, and the dimensional ratio and the like of each part may be different from the actual one and may be schematically illustrated. In addition, the description of each embodiment can be applied to each other unless otherwise specified.

<Cell type estimation method and estimation device>
The method of estimating the cell type of the present invention, as described above, for the test cell, an acquisition step of acquiring a signal light acquired using a coherent anti-Stokes Raman scattering (CARS) microscope, and based on the signal light, An estimation step of estimating the cell type of the test cell. Further, the cell type estimating device of the present invention is, based on the signal light, an acquisition unit that acquires signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell, and Estimating means for estimating the cell type of the cell. The estimation method of the present invention is characterized by estimating the cell type of the test cell based on the signal light acquired by the CARS microscope, and other configurations and conditions are not particularly limited.

As a result of earnest research, the present inventors have found that the signal light obtained by using a CARS microscope, specifically, the shape of the signal light, the signal intensity, and the like are related to the cell type. Came to be established. Therefore, according to the present invention, the cell type of the test cell can be estimated based on the signal light obtained by using the CARS microscope for the test cell. Moreover, the signal light from the CARS microscope can be obtained in a state where cells are alive. Therefore, according to the present invention, the cell type can be estimated in a state where the target cell, that is, the test cell is alive. Further, in the present invention, since the cell type is estimated based on the signal light obtained by using the CARS microscope, for example, labeling with a binding molecule such as an antibody specific to the cell surface marker molecule is unnecessary. Furthermore, according to the present invention, since it is possible to estimate the cell type, it is possible to estimate the differentiation stage or differentiation state of the cell, for example. Therefore, according to the present invention, for example, when a differentiated cell is induced from a pluripotent cell, progenitor cell, or other cell having differentiation potential, the differentiation stage or differentiation state of the obtained cell is estimated. be able to. Therefore, the estimation method of the present invention can also be referred to as, for example, a method of estimating the differentiation stage or differentiation state of cells. In the following, unless otherwise specified, the “differentiation state” has the same meaning as the “cell type” or the “differentiation stage”, and the explanations can be applied to each other by replacing them.

(Embodiment 1)
FIG. 1 shows a block diagram of the estimation device in this embodiment. As shown in FIG. 1, the estimation device 10 of this embodiment includes an acquisition unit 111 and an estimation unit 112. As shown in FIG. 1, the acquisition unit 111 and the estimation unit 112 may be incorporated in the data processing unit (data processing device) 11 that is hardware, or may be software or hardware in which the software is incorporated. The data processing means 11 may include a CPU or the like. Further, the data processing means 11 may include, for example, a ROM, a RAM, etc. described later.

Next, FIG. 2 illustrates a block diagram of a hardware configuration of the estimation device 10. The estimation device 10 includes, for example, a CPU (central processing unit) 201, a memory 202, a bus 203, a storage device 204, an input device 206, a display 207, a communication device 208, and the like. The respective units of the estimation device 10 are connected via a bus 203 by respective interfaces (I/F). The hardware configuration of the estimation apparatus 10 can be adopted as, for example, an estimation unit in an observation apparatus, an estimation unit in a manufacturing apparatus, and a hardware configuration of a learned model manufacturing apparatus described later.

The CPU 201 cooperates with other configurations by, for example, a controller (system controller, I/O controller, etc.) and controls the entire estimation apparatus 10. In the estimation device 10, the CPU 201 executes, for example, the program 205 of the present invention or other programs, and also reads or writes various information. Specifically, for example, the CPU 201 functions as the acquisition unit 111 and the estimation unit 112. The estimation device 10 includes a CPU as an arithmetic device, but may include other arithmetic devices such as a GPU (Graphics Processing Unit) and an APU (Accelerated Processing Unit), or may include a CPU and a combination thereof. Good. The CPU 201 functions as, for example, each unit other than the storage unit, the estimation unit, or the control unit in Embodiments 2 to 4 described later.

The memory 202 includes, for example, a main memory. The main memory is also called a main storage device. When the CPU 201 performs processing, the memory 202 reads various operation programs such as the program 205 of the present invention stored in the storage device 204 (auxiliary storage device) described later, for example. Then, the CPU 201 reads the data from the memory 202, decodes the data, and executes the program. The main memory is, for example, a RAM (random access memory). The memory 202 further includes, for example, a ROM (read-only memory).

The bus 203 can be connected to an external device, for example. Examples of the external device include an external storage device (external database and the like), a printer, and the like. The estimating apparatus 10 can be connected to a communication line network by the communication device 208 connected to the bus 203, for example, and can also be connected to the external device via the communication line network. The estimating apparatus 10 can also be connected to a terminal or the like via the communication device 208 and the communication network.

The storage device 204 is also called a so-called auxiliary storage device with respect to the main memory (main storage device), for example. As described above, the storage device 204 stores an operation program including the program 205 of the present invention. The storage device 204 includes, for example, a storage medium and a drive that reads and writes the storage medium. The storage medium is not particularly limited, and may be, for example, an internal type or an external type, HD (hard disk), FD (floppy (registered trademark) disk), CD-ROM, CD-R, CD-RW, MO, Examples thereof include DVD, flash memory, and memory card, and the drive is not particularly limited. The storage device 204 may be, for example, a hard disk drive (HDD) in which the storage medium and the drive are integrated.

The estimation device 10 further includes, for example, an input device 206 and a display 207. The input device 206 is, for example, a pointing device such as a touch panel, a track pad, or a mouse; a keyboard; an imaging means such as a camera or a scanner; a card reader such as an IC card reader or a magnetic card reader; a voice input means such as a microphone; To be Examples of the display 207 include a display device such as an LED display and a liquid crystal display. In the first embodiment, the input device 206 and the display 207 are separately configured, but the input device 206 and the display 207 may be integrally configured like a touch panel display.

Next, an example of the process of the estimation device 10 of the present embodiment will be described based on the flowchart of FIG. 3 taking the case where signal light is acquired for a test cell using a CARS microscope as an example.

Prior to the processing of the estimation device 10, first, a test cell is observed using a CARS microscope to acquire a signal light. For the configuration of the CARS microscope, the configuration of Example 1 described later can be referred to. The test cell may be any cell, for example, a cell whose cell type is known or a cell whose cell type is unknown. When the estimation device 10 is used to estimate the differentiation stage or differentiation state of cells, the test cells include, for example, cells before differentiation induction, cells during differentiation induction, cells after differentiation induction, and the like.

The test cell is not particularly limited, and can be any cell such as a cell separated from a living body or a cultured cell. Specific examples of the test cells include pluripotent cells such as iPS cells and ES cells, three germ layer cells derived from the pluripotent cells, and differentiated cells of each tissue derived from the three germ layer cells. can give. Examples of the differentiated cells include epithelial cells; neural cells such as neurons, glial cells and astrocytes; liver cells; adipocytes; stromal cells; hematopoietic cells such as T cells and B cells; skeletal muscle cells, myocardium Muscle cells such as cells; and the like. Examples of the three germ layer cells include ectoderm cells, mesoderm cells, and endoderm cells. According to the present invention, for example, which of the pluripotent cells, the ectodermal cells, the mesodermal cells, and the endodermal cells can be suitably estimated. As the markers for the pluripotent cells, the ectodermal cells, the mesodermal cells, and the endodermal cells, for example, the following markers can be used.
iPS cells: OCT4 (POU5F1), NANOG, DPPA4, DNMT3B, L1TD1, E-cadherin, Podocalyxin, TERT
Ectoderm cells: PAX6, SOX1, OTX2, MAP2, Musashi1, N-cadherin, Nestin, NCAM, GFAP
Mesodermal cells: T (Brachyury), Pitx2, MSX1, MESP1, GATA4, TBX6, NKX2.5, Vimentin, Desmin, Smooth Muscle Actin
Endoderm cells: SOX17, MIXL1, FOXA2, PDX1, AFP, CXCR4

Acquisition of signal light by the CARS microscope is performed for all or part of the test cells. In the latter case, it is preferable that the signal light is acquired by the CARS microscope, for example, in the central portion of the test cell, the signal light of a cross section including a direction perpendicular to the arrangement surface of the test cell is acquired. The number of test cells may be one or more. In the latter case, the signal light is obtained by the CARS microscope for some or all of the cells. The CARS microscope is a microscope that uses coherent anti-Stokes Raman scattering, and the configuration thereof is not particularly limited, and for example, the configurations of Examples described later can be referred to. The CARS microscope includes a light irradiating means for CARS spectroscopy, and the light irradiating means irradiates the test cell with the mixed light of the ultra-wide band light and the excitation light. Then, in the CARS microscope, CARS light (signal light) is dispersed from the emitted light emitted from the test cell by a diffraction grating or the like. Then, information such as the signal intensity of the CARS light and the spatial position is acquired. In the present invention, the signal light may include a second harmonic (SHG), a third harmonic (THG), and two-photon excitation fluorescence. The wavelength of the ultra wideband light is, for example, 400 to 2400 nm. The wavelengths of the excitation light are, for example, 750 to 1100 nm and 750 to 1064 nm.

Next, the processing by the estimation device 10 is started. First, the acquisition unit 111 of the estimation device 10 acquires the signal light, more specifically, the information of the signal light acquired by the CARS microscope (S1, acquisition step). The signal light acquired in the acquisition step includes signal light used in the estimation step described below, and as a specific example, includes at least one of CH ₂ expansion and contraction signal light and second harmonic (SHG) signal light. Is preferred. The signal light to be acquired can be appropriately set according to the molecule to be detected with reference to, for example, Tables 1A and 1B below. The Raman shift values shown in Tables 1A and 1B below are typical values, and the molecule to be detected can also be detected in the regions before and after the CARS microscope. Examples of the CH ₂ stretching signal light include CH ₂ symmetrical stretching vibration signal light, CH ₂ scissors bending vibration signal light, and the like. The wave number of the CH ₂ expansion/contraction signal light is, for example, 2835 to 2865 Raman shift/cm ⁻¹ . The wave number of the signal light of the CH ₂ symmetric stretching vibration is, for example, 2835 to 2865 Raman shift/cm ⁻¹ . The wave number of the signal light of the CH ₂ scissors bending vibration is, for example, 1423 to 1453 Raman shift/cm ⁻¹ . The wave number of the signal light of the SHG is λ/2, where λ is the wavelength of the incident laser.

Next, the estimation means 112 estimates the cell type of the test cell based on the signal light (S2, estimation step). In the signal light, each wave number (Raman shift/cm ⁻¹ ) is different in the substance in the test cell from which the signal is derived. Further, when the cell types are different, it is presumed that the cells having different cell types have different functions and metabolisms, so that the types of substances contained in the cells, the content rates of the substances, the localization of the substances and the like are also different. Therefore, in the present invention, the fact that the difference in the cell type of the cells occurs as the difference in the signal light is used to estimate the cell type of the test cell based on the signal light. When there are a plurality of test cells, the cell type of one or more test cells is estimated in the step S2, but it is preferable to estimate the cell types of all the test cells. Hereinafter, the test cell will be described with reference to an example of estimating whether the pluripotent cell, the ectodermal cell, the mesodermal cell, or the endodermal cell, the present invention However, the present invention is not limited to this, and can be used for estimating cells of other cell types.

The pluripotent cells and the ectodermal cells have abundant lipid droplets compared with, for example, the mesodermal cells and the endodermal cells. The fat droplets are mainly composed of triacylglycerol. Further, the fatty acid in triacylglycerol can be specified based on the CH ₂ stretching signal light. Therefore, the pluripotent cell and the ectodermal cell and the mesodermal cell and the endodermal cell are, for example, a signal light of CH ₂ stretching in fatty acid, and more preferably CH ₂ symmetric stretching vibration. It can be estimated by the signal light or the signal light of CH ₂ scissors bending vibration. Specifically, in the step S2, the pluripotent cells and the ectodermal cells and the mesodermal cells and the endodermal cells are, for example, CH ₂ stretch signal light, more specifically, , A reference value is set between the two based on the signal intensity of the CH ₂ expansion/contraction signal light, and the former can be estimated if the reference value is greater than or equal to the latter, and the latter can be estimated if the value is less than the reference value. The reference value can be set based on the signal intensity of the pluripotent cells, the ectodermal cells, the mesoderm cells, and the endoderm cells obtained by a CARS microscope in advance, for example. The signal intensity is preferably a corrected signal intensity corrected by the signal intensity of the reference sample (hereinafter the same). Examples of the reference sample include plastic beads such as polystyrene beads. The signal intensity is, for example, an average value of signal intensity in the acquired visual field.

S2 In step, based on the signal intensity of the signal light CH ₂ stretching, the shape of the CH ₂ stretching image reconstructed from the signal light of the CH ₂ stretching, pluripotent cells, ectodermal cells, mesodermal cells , And endoderm cells may be inferred. As a specific example, when a CH ₂ stretch image (lipid droplet) is present in a wide range of the cytoplasm and a region (cell nucleus) lacking a large signal is present in the CH ₂ stretch image (lipid droplet), in the S2 step, It can be presumed that the test cell is, for example, a pluripotent cell. When the CH ₂ stretch image (fat droplets) is present in a wide range of the cytoplasm and the CH ₂ stretch image (fat droplets) has a region (cell nucleus) lacking a moderate signal, in the S2 step, The test cells can be presumed to be ectodermal cells, for example. Furthermore, when granular CH ₂ stretch images (fat droplets) with weak signal intensity are present, it can be estimated that, in the S2 step, the test cells are, for example, mesodermal cells. When a granular CH ₂ stretch image (lipid droplet) with strong signal intensity is present, it can be estimated that the test cell is, for example, an endoderm cell in step S2. The shape can be detected using, for example, a shape detection method such as template matching.

Further, the ectodermal cells and the mesodermal cells have relatively strong signal intensity of SHG signal light as compared with the pluripotent cells and the endodermal cells. In the SHG image reconstructed from the SHG signal light of the ectodermal cells, the ectodermal cells generate, for example, a granular SHG signal (signal light or signal region, hereinafter the same). In the SHG image reconstructed from the SHG signal light of the pluripotent cells, the pluripotent cells generate, for example, filamentous SHG signals. On the other hand, in the SHG image reconstructed from the signal light of SHG of the mesodermal cells, the mesodermal cells generate, for example, reticulated SHG signals. In the SHG image reconstructed from the signal light of SHG of the endoderm cells, the endoderm cells do not generate the signal of SHG, for example. Therefore, the ectodermal cells and the mesoderm cells, and the pluripotent cells and the endodermal cells are, for example, based on the shape and/or the signal intensity of the signal light of SHG. You can presume to belong. Specifically, when the signal intensity of SHG signal light is used for estimation, in the step S2, the ectodermal cells and the mesodermal cells, and the pluripotent cells and the endodermal cells are, for example, A reference value is provided between the two, and it can be estimated that the value is equal to or more than the reference value and the value is less than the reference value. Further, when the shape of the SHG signal light is used for estimation, if the SHG signal light exists and the shape of the SHG (the signal) in the SHG image reconstructed from the SHG signal light is a filament shape, S2 In the process, it can be presumed to be pluripotent cells. When the signal light of SHG exists and the shape of (signal of) SHG is granular, it can be estimated that the cells are ectodermal cells in the step S2. Furthermore, when the signal light of SHG is present and the shape of (signal of) SHG is reticulated, it can be estimated that it is a mesodermal cell in the step S2. When the signal light of SHG does not exist, it can be estimated that the cells are endoderm cells in step S2.

In the S2 step, based on the signal intensity of the SHG signal light and the shape of the SHG signal in the SHG image reconstructed from the SHG signal light, pluripotent cells, ectodermal cells, mesodermal cells, and Endoderm cells may be inferred. As a specific example, when a filamentous SHG image (SHG signal) is present, in the S2 step, the test cell can be estimated to be, for example, a pluripotent cell. Further, when a granular SHG image (signal of SHG) having a strong signal intensity is present, in the S2 step, the test cell can be estimated to be, for example, an ectodermal cell. Further, when a mesh-shaped (reticular) SHG image (signal of SHG) is present, it can be estimated in the step S2 that the test cell is, for example, a mesodermal cell. Then, when the SHG image (SHG signal) does not exist, it can be estimated that the test cell is, for example, an endoderm cell in the step S2.

In step S2, the estimation means 112 may combine a plurality of conditions to estimate the cell type. Specifically, in the step S2, it is determined whether or not the signal intensity of the CH ₂ expansion/contraction signal light in the signal light is equal to or higher than a reference value. When the signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than the reference value, in the step S2, it is determined whether the signal intensity of the second harmonic (SHG) signal light in the signal light is equal to or higher than the reference value. judge. Then, when the signal intensity of the SHG signal light in the signal light is equal to or higher than a reference value, it is estimated that the test cell is an ectodermal cell in step S2. On the other hand, when the signal intensity of the signal light of SHG in the signal light is less than the reference value, it is estimated in the step S2 that the test cell is a pluripotent cell. In the step S2, the cell tumor may be estimated based on the shape of the SHG signal in the SHG image reconstructed from the signal light instead of or in addition to the signal intensity of the SHG signal light in the signal light. Specifically, in step S2, it may be determined whether the shape of the SHG signal in the SHG image is granular or filamentous. Then, when the shape of the SHG signal in the SHG image is granular or not filamentous, it is presumed that the test cells are ectodermal cells in step S2. On the other hand, when the shape of the SHG signal in the SHG image is not granular or is filamentous, it is presumed that the test cells are pluripotent cells in step S2.

Next, when the signal intensity of the CH ₂ expansion and contraction signal light in the signal light is less than the reference value, it is determined in step S2 whether the signal intensity of the SHG signal light in the signal light is equal to or more than the reference value. To do. Then, when the signal intensity of the signal light of SHG in the signal light is equal to or higher than the reference value, it is estimated in the step S2 that the test cell is a mesodermal cell. On the other hand, when the signal intensity of the signal light of SHG in the signal light is less than the reference value, it is estimated that the test cell is an endoderm cell in step S2. In the step S2, a cell tumor may be estimated based on the shape of the SHG signal in the SHG image instead of or in addition to the signal intensity of the SHG signal light in the signal light. Specifically, in the step S2, it may be determined whether or not the shape of the SHG signal in the signal light is reticulated or the shape exists. Then, when the shape of the SHG signal in the SHG image is reticular or when there is a shape, the test cell is estimated to be a mesodermal cell in step S2. On the other hand, when the shape of the SHG signal in the SHG image is not reticulated or does not exist, it is presumed that the test cells are endoderm cells in step S2.

As a specific example of the plurality of conditions in the step S2, when estimating whether the condition is the pluripotent cell, the ectodermal cell, the mesodermal cell, or the endodermal cell, in the S2 step, for example, Based on the signal light and the following conditions (1) to (4), it can be determined whether the signal light satisfies the following conditions (1) to (4), and the cell type of the test cell can be estimated.
(1) When the signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value and the signal intensity of the second harmonic signal light in the signal light is equal to or lower than the reference value (less than), or When the shape of the signal light of the second harmonic in the signal light is filamentous, it is estimated that the test cell is a pluripotent cell.
(2) The signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or higher than the reference value, or in the signal light When the signal light of the second harmonic has a granular shape, it is estimated that the test cell is an ectodermal cell.
(3) The signal intensity of the CH ₂ expansion/contraction signal light in the signal light is less than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or greater than the reference value, or in the signal light When the shape of the second harmonic signal light is reticulated, it is estimated that the test cell is a mesodermal cell.
(4) When the signal intensity of the CH ₂ expansion and contraction signal light in the signal light is less than a reference value and the signal intensity of the second harmonic signal light in the signal light is less than the reference value, the test cell Are presumed to be endodermal cells.

The pluripotent cells and the ectodermal cells have relatively large cell nuclei, for example, as compared with the mesodermal cells and the endodermal cells. Therefore, in the step S2, the size of the cell nucleus may be used instead of the CH ₂ expansion and contraction in the signal light under the conditions (1) to (4). The size of the cell nucleus is represented, for example, as the size of a region having a weak signal intensity in a CH ₂ stretch image reconstructed from CH ₂ stretch signal light.

In step S2, the estimation unit 112 uses, for example, a learned model generated by machine learning that outputs an estimation result of the cell type of the test cell based on the signal light, and uses the learned cell type of the test cell. May be estimated. The learned model can be manufactured by, for example, a learned model manufacturing method and a learned model manufacturing apparatus described later.

In the first embodiment, the estimation unit 112 detects the signal intensity and the shape of the signal light of the second harmonic, but for example, the detection unit that detects the signal intensity of each signal light or the signal image from each signal light. May be reconstructed and detected by a detecting means for detecting the shape.

Although the pluripotent cells, ectodermal cells, mesoderm cells, and endoderm cells have been described as examples of the test cells in Embodiment 1, the present invention is not limited thereto as described above. Instead, it can be used to estimate other cells. As a specific example, when the SHG image reconstructed from SHG is axonal, the test cell can be estimated to be, for example, a nerve cell. Moreover, when the SHG image reconstructed from SHG has a pili root shape, the test cell can be estimated to be, for example, a photoreceptor cell.

In Embodiment 1, the case where CH ₂ expansion and contraction and SHG are used as the signal light has been described as an example, but other signal light may be used. As a specific example, when detecting bone-type cells as the test cells, PO ₄ ³⁻ can be detected by using 961 Raman shift/cm ⁻¹ as the signal light.

<Cell manufacturing method and manufacturing apparatus>
As described above, the method for producing cells of the present invention comprises an observation step of observing a test cell, an estimation step of estimating a cell type of the test cell, and a test cell of the predetermined cell type. Including a selection step, the observation step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope, and the estimation step is performed by the cell type estimation method of the present invention. Further, the cell manufacturing apparatus of the present invention is an observation unit capable of observing a test cell, an estimation unit for estimating a cell type of the test cell, and a selection unit for selecting a test cell of the predetermined cell type. And the observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope, and the estimation unit includes the cell type estimation device of the present invention. The production method and production apparatus of the present invention are characterized in that the estimation of the cell type of the test cell is performed by the estimation method or estimation apparatus of the present invention, and other configurations and conditions are not particularly limited. According to the present invention, a cell type can be estimated in a living state of a target cell, and thus a cell of a predetermined cell type can be produced in a living state. The description of the estimation method and the estimation apparatus of the present invention can be applied to the production method and the production apparatus of the present invention.

(Embodiment 2)
FIG. 4 shows a block diagram of the manufacturing apparatus in this embodiment. As shown in FIG. 4, the manufacturing apparatus 20 of the present embodiment includes the estimation device 10 of the first embodiment as an estimation unit, and further includes an observation unit 113 and a selection unit 114. The observation unit 113 can observe the test cells. The observation unit 113 is, for example, an optical microscope such as a bright field microscope, a stereomicroscope, a phase contrast microscope, a differential interference microscope, a polarization microscope, a fluorescence microscope, a confocal laser microscope, a total reflection illumination fluorescence microscope, a Raman microscope, and a CARS microscope. To be The observation unit 113 preferably includes a CARS microscope. In this case, the observation unit 113 may include another optical microscope. The selection unit 114 selects test cells of a predetermined cell type. The selection unit 114 may be, for example, capable of collecting or maintaining a test cell of a predetermined cell type, and specific examples thereof include a laser irradiation unit capable of irradiating the test cell with a laser, a recovery unit capable of recovering a specific cell, and the like. can give. In the former case, the laser irradiation unit may, for example, irradiate the test cells of the predetermined cell type with a laser and collect them by peeling, or irradiate the cells other than the predetermined cell type with a laser. Alternatively, it may be peeled off or killed. In the latter case, the recovery unit may be, for example, a combination of a scalar robot and a scraper. Except for these points, the manufacturing apparatus 20 of the second embodiment has the same configuration as the estimation apparatus 10 of the first embodiment, and the description thereof can be incorporated.

Next, an example of processing in the manufacturing apparatus of FIG. 4 will be described based on the flowchart of FIG.

First, the test cell is observed by the observation unit 113 (S3, observation step). Thereby, the signal light acquired by the CARS microscope necessary for estimating the cell type of the test cell is acquired. The step S3 can be carried out in the same manner as the acquisition of the signal light by the CARS microscope in the first embodiment.

Next, in the same manner as the steps S1 and S2 in the estimation method of the first embodiment, the estimation device 10 that is an estimation unit estimates the cell type of the test cell (estimation step).

Further, the selection unit 114 selects test cells of a predetermined cell type (S4, selection step). The predetermined cell type is not particularly limited and can be any cell type. The selection by the selection unit 114 may be performed, for example, by collecting test cells of a predetermined cell type or by removing test cells other than the predetermined cell type. As a specific example, in the case of inducing a specific differentiated cell from the pluripotent cell, the predetermined cell type is, for example, a specific differentiated cell. Therefore, in the step S4, the selection unit 114 collects the specific differentiated cells and removes cells before the specific cell type, so that the specific differentiated cells can be selected.

<Observation method and observation device>
The observation method using the coherent anti-Stokes Raman scattering (CARS) of the present invention includes, as described above, an observation step of observing a test cell, an estimation step of estimating the cell type of the test cell, and a predetermined cell. Re-observing the test cells of the species again, the observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope, and the estimating step is the method of estimating the cell type of the present invention. It is carried out by. An observation apparatus using coherent anti-Stokes Raman scattering (CARS) of the present invention is capable of observing a test cell, an estimation unit for estimating a cell type of the test cell, and the control of the observation unit. And a control unit, the observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope, the estimation unit includes the cell type estimation device of the present invention, and the control unit includes one of the test cells. The cells presumed to be a predetermined cell type are re-observed by the observation unit. The observation method and the observation device of the present invention are characterized in that the estimation of the cell type of the test cell is performed by the estimation method or the estimation device of the present invention, and other configurations and conditions are not particularly limited. According to the present invention, since the cell type can be estimated in a living state of the target cell, it is possible to further observe a cell of a predetermined cell type in a living state. Regarding the observation method and the observation apparatus of the present invention, the description of the estimation method, the estimation apparatus, the manufacturing method, and the manufacturing apparatus of the present invention can be applied.

(Embodiment 3)
FIG. 6 shows a block diagram of the observation device in this embodiment. As shown in FIG. 6, the observation device 30 of this embodiment includes an observation unit 113 and an estimation unit 31. The estimation unit 31 includes an acquisition unit 111, an estimation unit 112, and a control unit 115. The control unit 115 can control the observation unit 113, and the observation unit 113 re-observes the cells estimated to be a predetermined cell type among the test cells. Except for these points, the observation device of the third embodiment has the same configuration as the estimation device 10 of the first embodiment or the manufacturing device 20 of the second embodiment, and the description thereof can be incorporated.

Next, an example of processing in the observation device in FIG. 6 will be described based on the flowchart in FIG. 7.

First, similarly to the manufacturing method of the second embodiment, the test cell is observed by the observation unit 113 (S3, observation step). Next, the cell type of the test cell is estimated based on the obtained signal light in the same manner as the steps S1 and S2 in the estimation method of the first embodiment (estimation step).

Further, the control unit 115 controls the observation unit 113, and the observation unit 113 again observes the test cells of a predetermined cell type (S5, re-observation step).

<Learning model manufacturing method and manufacturing device>
As described above, the manufacturing method of the learned model used for estimating the cell type of the present invention includes an acquisition step of acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of the test cell, A learning step of generating a learned model that outputs the estimation result of the cell type of the test cell from the signal light using the pair of the signal light and the cell type of the test cell as teacher data. Further, the trained model manufacturing apparatus used for estimating the cell type of the present invention includes an acquisition unit for acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell, and the signal light. And learning means for generating a learned model that outputs the estimation result of the cell type of the test cell from the signal light, using the set of the cell type of the test cell as teacher data. The learning method and the learning device of the present invention use a pair of a signal light obtained by using a CARS microscope for a test cell and a cell type of the test cell as teaching data, and from the signal light, a cell type of the test cell. The feature is that a trained model that outputs the estimation result of is generated, and other configurations and conditions are not particularly limited. According to the learning model manufacturing method and the manufacturing apparatus of the present invention, it is possible to manufacture a learned model capable of estimating the cell type in a state where the target cell is alive. Regarding the learning method and the learning device of the present invention, the description of the estimation method, the estimation device, the manufacturing method, the manufacturing device, the observation method and the observation device of the present invention can be applied.

(Embodiment 4)
FIG. 8 shows a block diagram of the learning device in this embodiment. As shown in FIG. 8, the learning device 40 of this embodiment includes an acquisition unit 111 and a learning unit 116. As shown in FIG. 8, the acquisition unit 111 and the learning unit 116 may be incorporated in the data processing unit (data processing device) 11 that is hardware, or may be software or hardware in which the software is incorporated. The data processing means 11 may include a CPU or the like. Further, the data processing means 11 may include, for example, the above-mentioned ROM, RAM and the like. The acquisition unit 111 is the same as the acquisition unit 111 in the estimation device 10 of the first embodiment, and the description thereof can be applied.

Next, an example of processing in the learning device in FIG. 8 will be described based on the flowchart in FIG.

First, similarly to step S1 of the estimation method of the first embodiment, the acquisition unit 111 acquires the signal light acquired using the CARS microscope. The number of the test cells is not particularly limited and can be any number. When there are a plurality of test cells, the cell type of each test cell may be the same or different. In the latter case, it is preferable that some test cells have overlapping cell types. As a specific example, when a learned model capable of estimating the pluripotent cells, the ectodermal cells, the mesodermal cells and the endodermal cells is generated, the test cells are cell types. It is preferable to use the confirmed pluripotent cells, the ectodermal cells, the mesodermal cells, and the endodermal cells. Also, each cell is preferably plural.

Next, the learning unit 116 generates a learned model that outputs the estimation result of the cell type of the test cell from the signal light, using the pair of the signal light and the cell type of the test cell as teaching data. (S6, learning process). Specifically, for each test cell, the signal light obtained in step S1 is associated with the cell type of the test cell. The associated signal light may be the entire spectrum of the signal light obtained in step S1 or a specific signal light. In the latter case, the specific signal light includes, for example, CH ₂ expansion/contraction signal light and SHG signal light. In the case of associating the specific signal light, the learning method according to the present embodiment detects the specific signal light in the obtained signal by the detecting unit after step S1, for example. Further, for example, the detection unit may detect the signal intensity of the specific signal light, or may detect the shape of the signal light in the signal image obtained by reconstructing the specific signal light.

The machine learning method used to generate the learned model is not particularly limited, and for example, the learning technique used for classification can be used. As a specific example, examples of the machine learning method include a support vector machine, an extreme learning machine, and a feature learning. In the present embodiment, the machine learning uses supervised learning, but semi-supervised learning may be used. The number of the teacher data is plural, and the upper limit is not particularly limited.

<Program>
A program of the present invention is a program capable of executing the estimation method, manufacturing method, observation method or learning method of the present invention on a computer. Alternatively, the program of this embodiment may be recorded in a computer-readable recording medium, for example. The recording medium is, for example, a non-transitory computer-readable storage medium. The recording medium is not particularly limited, and examples thereof include a random access memory (RAM), a read-only memory (ROM), a hard disk (HD), an optical disk, and a floppy (registered trademark) disk (FD).

Next, an embodiment of the present invention will be described. However, the present invention is not limited to the following examples. Commercially available reagents were used according to their protocol unless otherwise indicated.

[Example 1]
It was confirmed that the signal light obtained by the CARS microscope has a relationship with the cell type of the cell.

(1) CARS Microscope The CARS microscope used in Example 1 was a multiplex CARS microscope. A schematic diagram of the configuration of the optical system of the CARS microscope used in Example 1 is shown in FIG.

As the light source, a cw Q switch microchip Nd:YAG laser (A) having an oscillation wavelength of 1064 nm, a pulse width of 800 ps and a repetition frequency of 33 kHz was used. The pulse width is sub-nanosecond and has a line width of about 1 cm ^-1 or less. Divide the output into two, a pulse laser having a center wavelength of 1064nm is fundamental as one is omega ₁ light and the other is introduced into PCF, it was generated supercontinuum light as omega ₂ light. The ω ₁ light is the laser light emitted from the light source body with a plano-convex lens (B) of f=400.0 mm (AC254-400-C f=400.0 mm, φ1" Achromatic Doublet, ARC: 1050-1700 nm; Thorlabs). The ω ₂ light is collimated and the ω ₂ light is a super-continuum light having a broadband wavelength component emitted from the PCF. ) (C) was introduced and collimated and propagated.

Further, the ω ₁ light was sandwiched between a Variable Neutral Density Filter (VND Filter) (D) and a 1064 nm half-wave plate (S333-1064-2; manufactured by Suruga Seiki Co., Ltd.) (E) on the way of propagation. The VND filter was used to adjust the light amount of ω ₁ light incident on the sample, and the 1064 nm half-wave plate was used to confirm the polarization dependence of SHG active molecules described later.

Furthermore, the ω ₁ light contains a weak component on the short wavelength side in addition to the central wavelength of 1064 nm, and in order to efficiently cut the extra spectral components contained in the ω ₁ light other than the CARS light to be detected. A 1064 nm narrow band pass filter (1064.1-1 OD7 Ultra Narrow Bandpass; Alluxa) (F) was used. The bandpass filter used was an ultra-narrow bandpass filter that transmits only a spectral component having a center wavelength of 1064.1 nm and a half-value width of 1 nm.

ω ₂ light has a broad band that slowly extends from the visible region to the near infrared region by PCF. The wavelength range of ω ₂ light required for measuring CARS light is about 1064 to 1650 nm with respect to the central wavelength of 1064 nm of ω ₁ light. Accordingly, two filters 1050 nm cuts components of visible light immediately after collimation of the omega ₂ light long Pass Filter (G), infrared transmission filter (IR80N; Kenko manufactured optical Ltd.) (H) near red omega ₂ light by Only the spectral components that spread to the outer region were sufficiently transmitted. Then, after combining the ω ₁ light and the ω ₂ light, the two lights were propagated to the microscope.

As the microscope, an upside down microscope (eclipse Ti-U, manufactured by Nikon Co., Ltd.) that is custom-made is used. Nonlinear optics such as CARS and SHG by ω ₁ light and ω ₂ light incident on the objective lens (Water immersion, Plan, ×60, 1.27NA; Nikon) (J) from the inverted side of the microscope and focused on the sample. The phenomenon occurs. Since they are focused by the objective lens, these nonlinear optical phenomena efficiently occur in the forward direction due to the phase matching condition. The signal light generated at the focal point was condensed and collimated by an objective lens (Dry, S Plan Fluor, ×40, 0.60NA; manufactured by Nikon) on the upright side of the microscope (K).

The average power measured with a power meter at the position where the light was focused on the sample surface by the inverted objective lens was ω ₁ light of maximum 55 mW and ω ₂ light of maximum 20 mW, respectively. In the actual measurement, the sample signal intensity was adjusted by using the above-mentioned VND filter for the ω ₁ light with high power.

The microscope is equipped with a halogen lamp, and an optical image of the sample can be taken with a CCD camera. In the microscope, ω ₁ light and ω ₂ light exist coaxially with the light of the halogen lamp, so the dichroic beam splitter (FF825-SDi01-25×36×2.0 single edge short path Dichroic beam splitter; opt (Manufactured by Line Co., Ltd.) (L) and a dichroic mirror (TFMS-30C05-3/20 ultra-wide band induction multilayer flat mirror; manufactured by Sigma Optical Co., Ltd.) (M) immediately after the erecting side objective lens were used.

The stage installed in the microscope used a step motor type MicroStage (Micro-Stage(2 axis); MadCityLabs)(N) controlled by a biaxial micro-positioning device. Furthermore, a piezo stage (Nano-LPQ; made by Mad City Labs) (O) with an operating range of 75 μm × 75 μm × 50 μm (O) is installed on the MicroStage, and in addition to a wide in-plane stroke in millimeters In-plane stage control of the stroke and scanning in the optical axis Z direction, that is, in the depth direction with respect to the sample are enabled. MicroStage can control a minimum step size of 95nm and a maximum speed of 2mm/sec with an extremely high precision drive.

Next, the optical system on the detector side was as follows. The signal light collected by the erecting objective lens of the microscope is reflected by the dichroic mirror immediately after, and then the dichroic beam splitter (FF685-Di02-25×36 685 nm  single-edge Dichroic beam splitter; manufactured by Semrock) (P ) Transmitted near-infrared CARS light and reflected visible light.

For light in the visible range, the 633 nm short pass filter (Q) blocks the background from excitation light outside the visible range, and then a plano-convex lens (R) with f=65 mm (Z-300 Series; manufactured by LUCIR) ) (S). Control was performed by software (Ementool; Zolix) on a PC connected to the spectroscope. In this embodiment, the grating Groove is set to 300 lines/mm, the grating Blaze is set to 500 nm, and the center wavelength of the spectroscope is set to 410 nm. An electronic cooling CCD camera (iVac300; manufactured by Andor) (T) was used to detect SHG.

Light in the near-infrared region passes through the dichroic beam splitter, and ω ₁ light and ω ₂ light are transmitted by the 1064nm notch filter (1064 Narrow notch filter; manufactured by Iridian) (U) and the 1050nm short pass filter (3RD1050SP; Omega) (V). Was cut off, and only CARS light was guided to the spectroscope.

After passing through the 1064 nm notch filter and 1050 nm short pass filter, the CARS light is spectrally separated by wavelength with a spectroscope (Acton Series LS785; Princeton Instruments) (X) with a plano-convex lens of f=25 mm, and electronically cooled CCD camera (PIXIS 100BR; manufactured by Princeton Instruments) (Y). The PIXIS100BR, iVac316, and iVac300 are controlled by software (LightField; Princeton Instruments, AndorSOLIS; Andor) on the PC.

(2) Test cells As test cells, iPS cells (HiPS-WTc11 (GM25256)) and ectodermal cells, mesodermal cells, and endodermal cells derived from the iPS cells are used. I was there. For each cell, a preparation was placed in a 24-well dish and cultured on the preparation. Each cell was prepared as follows. For iPS cells, the cells seeded on the plate were collected and seeded again at 2500 cells/cm ² . Stem Fit AK02N (manufactured by Ajinomoto Co., Inc.) supplemented with Supplement B and Supplement C was used for culturing iPS cells. For ectodermal cells, mesodermal cells, and endodermal cells, iPS cells were seeded at 2500 cells/cm ² , cultured for 1 day, and then replaced with Stem Fit AK02N medium containing no Supplement C. did. At this time, when inducing ectodermal cells, 10 μmol/L SB431542 (4-[4-(1,3-benzodioxol-5-yl)-5-(2-pyridinyl)-1H-imidazole) was added to the medium. -2-yl]benzamide (ALK inhibitor), Wako Pure Chemical Industries, Ltd. and 10 μmol/L DMH1 (4-[6-(4-Isopropoxyphenyl)pyrazolo[1,5-a]pyrimidin-3-yl]quinoline, 4 -Add each compound so that it becomes -[6-[4-(1-Methylethoxy)phenyl]pyrazolo[1,5-a]pyrimidin-3-yl]-quinoline (BMP inhibitor, Wako Pure Chemical Industries) did. When inducing mesodermal cells, 3 μmol/L CHIR99021 (6-[[2-[[4-(2,4-Dichlorophenyl)-5-(5-methyl-1H-imidazol- 2-yl)-2-pyrimidinyl]amino]ethyl]amino]-3-pyridinecarbonitrile (GSK3 inhibitor), Wako Pure Chemical Industries, Ltd. and 10 ng/mL human recombinant Bone Morphogenetic Protein 4 (Wako Pure Chemical Industries, Ltd.) , Compound and protein were added. Furthermore, when inducing endoderm cells, the compound and the protein were added so that the concentration was 3 μmol/L CHIR99021 (Wako Pure Chemical Industries, Ltd.) and 10 ng/mL human recombinant Activin A (Wako Pure Chemical Industries, Ltd.). Then, each cell was cultured for 4 days or more.

(3) Measurement The preparation containing each cell after culturing was mounted on a slide glass, and signal light was acquired with the CARS microscope of Example 1(1). Then, the obtained CH ₂ stretching and SHG signal light was reconstructed, and an image of CH ₂ symmetric stretching vibration (CH ₂ stretching image) and an SHG image were prepared. The results are shown in FIG.

FIG. 11 is a photograph showing a CH ₂ stretch image and an SHG image. 11, (A) shows the results of iPS cells, (B) shows the results of ectodermal cells, (C) shows the results of mesodermal cells, and (D) shows the internal endoderm cells. The result of a germ cell is shown. Further, in FIG. 11, the upper part shows an SHG image, and the lower part shows a CH ₂ stretch image. As shown by the arrow X in the upper part of FIGS. 11A to 11D, as the SHG image, a filamentous SHG image is observed in iPS cells, and a granular SHG image with strong signal intensity in ectodermal cells. , A mesh-shaped (reticular) SHG image was observed in the mesodermal cells, and no SHG image was observed in the endodermal cells. Therefore, it was found that the cell type of each cell can be estimated based on the signal intensity and shape of the SHG image. Further, as shown in the lower part of FIGS. 11A to 11D, as the CH ₂ stretch image, the CH ₂ stretch image was observed in a wide range in the iPS cells, and the large signal was absent in the CH ₂ stretch image. the region (nucleus) is observed, the ectodermal cells, a wide range CH ₂ stretching image is observed in, and the area of lack of moderate signal in the CH ₂ stretching images (cell nuclei) were observed, mesoderm the system cells, are CH ₂ stretching images granular observation, the endodermal cells, CH ₂ stretching image signal intensity is strong graininess was observed. Therefore, it was found that the cell type of each cell can be estimated based on the signal intensity and shape of the CH ₂ stretch image.

From the above, it was found that the signal light obtained by the CARS microscope has a relationship with the cell type, and the cell type can be estimated based on the signal light.

Although the present invention has been described with reference to the exemplary embodiments and examples, the present invention is not limited to the above exemplary embodiments and examples. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

This application claims priority based on Japanese Patent Application No. 2019-022032 filed on February 8, 2019, and incorporates all of the disclosure thereof.

<Appendix>
The whole or part of the exemplary embodiments and examples described above can be described as, but not limited to, the following supplementary notes.
(Appendix 1)
An acquisition step of acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
An estimation step of estimating the cell type of the test cell based on the signal light.
(Appendix 2)
The estimation method according to appendix 1, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
(Appendix 3)
The estimation method according to appendix 2, wherein the CH ₂ stretching signal light is CH ₂ symmetrical stretching vibration signal light or CH ₂ scissors bending vibration signal light.
(Appendix 4)
4. The estimation method according to any one of appendices 1 to 3, wherein in the estimation step, the cell type of the test cell is estimated based on at least one of the shape and signal intensity of the CH ₂ stretching signal light in the signal light.
(Appendix 5)
The estimation method according to any one of appendices 1 to 4, wherein in the estimation step, the cell type of the test cell is estimated based on at least one of the shape and signal intensity of the second harmonic signal light in the signal light. ..
(Appendix 6)
6. In the estimation step, it is estimated whether the cell type of the test cell is a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell, in any one of appendixes 1 to 5. Estimation method.
(Appendix 7)
7. The estimation method according to any one of appendices 1 to 6, wherein in the estimation step, the cell type of the test cell is estimated based on the signal light and the following conditions (1) to (4):
(1) When the signal intensity of the CH ₂ expansion/contraction signal light in the signal light is a reference value or more and the signal intensity of the second harmonic signal light in the signal light is a reference value or less, the test cell Are presumed to be pluripotent cells;
(2) The signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or higher than the reference value, or in the signal light When the shape of the second harmonic signal light is granular, the test cell is presumed to be an ectodermal cell;
(3) The signal intensity of the CH ₂ expansion/contraction signal light in the signal light is less than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or greater than the reference value, or in the signal light When the shape of the second-harmonic signal light is reticulated, it is presumed that the test cell is a mesodermal cell; and (4) the signal intensity of the CH ₂ stretching signal light in the signal light is a reference. When it is less than the value and the signal intensity of the signal light of the second harmonic in the signal light is less than the reference value, it is estimated that the test cell is an endoderm cell.
(Appendix 8)
In the estimating step, the cell type of the test cell is estimated using a learned model generated by machine learning that outputs an estimation result of the cell type of the test cell based on the signal light. The estimation method according to any one of 1 to 7.
(Appendix 9)
9. The estimation method according to any one of appendices 1 to 8, wherein the test cell is a pluripotent cell or a differentiated cell derived from a pluripotent cell.
(Appendix 10)
Acquisition means for acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
A cell type estimation device comprising: an estimation unit that estimates the cell type of the test cell based on the signal light.
(Appendix 11)
The estimation device according to appendix 10, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
(Appendix 12)
The estimation device according to appendix 11, wherein the CH ₂ expansion/contraction signal light is CH ₂ symmetrical expansion/contraction vibration signal light.
(Appendix 13)
13. The estimation device according to any one of appendices 10 to 12, wherein the estimation unit estimates the cell type of the test cell based on at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light in the signal light. ..
(Appendix 14)
The estimation according to any one of appendices 10 to 13, wherein the estimation unit estimates the cell type of the test cell based on at least one of the shape and signal intensity of the signal light of the second harmonic in the signal light. apparatus.
(Appendix 15)
In any one of appendices 10 to 14, the estimation means estimates whether the cell type of the test cell is a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell. The estimation device described.
(Appendix 16)
16. The estimating device according to any one of appendices 10 to 15, wherein the estimating unit estimates the cell type of the test cell based on the signal light and the following conditions (1) to (4):
(1) When the signal intensity of the CH ₂ expansion/contraction signal light in the signal light is a reference value or more and the signal intensity of the second harmonic signal light in the signal light is a reference value or less, the test cell Are presumed to be pluripotent cells;
(2) The signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or higher than the reference value, or in the signal light When the shape of the second harmonic signal light is granular, the test cell is presumed to be an ectodermal cell;
(3) The signal intensity of the CH ₂ expansion/contraction signal light in the signal light is less than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or greater than the reference value, or in the signal light When the shape of the second-harmonic signal light is reticulated, it is presumed that the test cell is a mesodermal cell; and (4) the signal intensity of the CH ₂ stretching signal light in the signal light is a reference. When it is less than the value and the signal intensity of the signal light of the second harmonic in the signal light is less than the reference value, it is estimated that the test cell is an endoderm cell.
(Appendix 17)
In the estimation means, the cell type of the test cell is estimated using a learned model generated by machine learning that outputs an estimation result of the cell type of the test cell based on the signal light. The estimation device according to any one of 10 to 16.
(Appendix 18)
The estimation device according to any one of appendices 10 to 17, wherein the test cell is a pluripotent cell or a differentiated cell derived from the pluripotent cell.
(Appendix 19)
An observation step of observing the test cells,
An estimation step of estimating the cell type of the test cell,
And a selection step of selecting test cells of a predetermined cell type,
The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
The method for producing a cell, wherein the estimating step is performed by the method for estimating a cell type according to any one of appendices 1 to 9.
(Appendix 20)
20. The selection step selects test cells of the predetermined cell type by collecting test cells of the predetermined cell type or removing test cells other than the predetermined cell type. The method for producing cells.
(Appendix 21)
An observation unit capable of observing test cells,
An estimation unit for estimating the cell type of the test cell,
Including a selection unit for selecting test cells of a predetermined cell type,
The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
The said estimation unit is a cell manufacturing apparatus containing the estimation apparatus of the cell type in any one of appendixes 10-18.
(Appendix 22)
The selection unit is a collection means for collecting test cells of the predetermined cell type or test cells other than the predetermined cell type, or laser irradiation capable of irradiating the test cells other than the predetermined cell type with laser. 22. The cell manufacturing apparatus according to appendix 21, comprising a unit.
(Appendix 23)
An observation step of observing the test cells,
An estimation step of estimating the cell type of the test cell,
Including a re-observation step of observing the test cells of a predetermined cell type again,
The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
An observation method using a coherent anti-Stokes Raman scattering (CARS) microscope, wherein the estimation step is performed by the cell type estimation method according to any one of appendices 1 to 9.
(Appendix 24)
An observation unit capable of observing test cells,
An estimation unit for estimating the cell type of the test cell,
A control unit capable of controlling the observation unit,
The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
The estimation unit includes the cell type estimation device according to any one of appendices 10 to 18,
An observation apparatus using a coherent anti-Stokes Raman scattering (CARS) microscope, wherein the control unit re-observes cells estimated to be a predetermined cell type among the test cells by the observation unit.
(Appendix 25)
An acquisition step of acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell;
Cell type estimation including a learning step of generating a learned model that outputs an estimation result of the cell type of the cell from the signal light, using a pair of the signal light and the cell type of the test cell as teaching data. Manufacturing method of trained model used for.
(Appendix 26)
26. The manufacturing method according to appendix 25, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
(Appendix 27)
27. The manufacturing method according to appendix 26, wherein the CH ₂ stretching signal light is CH ₂ symmetrical stretching vibration signal light.
(Appendix 28)
28. The manufacturing method according to any one of appendices 25 to 27, wherein in the learning step, at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light is used as the signal light.
(Appendix 29)
29. The manufacturing method according to any one of appendices 25 to 28, wherein in the learning step, at least one of the shape and signal intensity of the second harmonic signal light is used as the signal light.
(Appendix 30)
30. The method according to any one of appendixes 25 to 29, wherein the test cell includes a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell.
(Appendix 31)
An acquisition means for acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of the test cell;
A cell type including a learning unit that generates a learned model that outputs an estimation result of the cell type of the test cell from the signal light, using a pair of the signal light and the cell type of the test cell as teaching data. Manufacturing device for trained model used for estimation.
(Appendix 32)
32. The manufacturing apparatus according to appendix 31, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
(Appendix 33)
33. The manufacturing apparatus according to appendix 32, wherein the CH ₂ stretching signal light is CH ₂ symmetrical stretching vibration signal light.
(Appendix 34)
34. The manufacturing apparatus according to any one of appendices 31 to 33, wherein the learning unit uses at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light as the signal light.
(Appendix 35)
35. The manufacturing apparatus according to any one of appendices 31 to 34, wherein in the learning means, at least one of the shape and the signal intensity of the second harmonic signal light is used as the signal light.
(Appendix 36)
36. The manufacturing apparatus according to any one of appendices 31 to 35, wherein the test cell includes a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell.
(Appendix 37)
An acquisition process for acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
A program capable of executing, on a computer, an estimation process of estimating the cell type of the test cell based on the signal light.
(Appendix 38)
An acquisition process for acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell;
On the computer, a learning process of generating a learned model that outputs an estimation result of the cell type of the test cell from the signal light using the pair of the signal light and the cell type of the test cell as teaching data. A program that is executable.
(Appendix 39)
A computer-readable recording medium recording the program according to appendix 38 or 39.

As described above, according to the present invention, it is possible to estimate the cell type while the target cell is alive. Therefore, the present invention is extremely useful in the fields of life science, processing of cells and tissues, regenerative medicine and the like.

10 Estimating device 11 Data processing means 111 Acquisition means 112 Estimating means 113 Observation unit 114 Selection unit 115 Control unit 116 Learning means 20 Manufacturing device 30 Observation device 31 Estimating unit 40 Learning device

Claims (38)

1. An acquisition step of acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
   An estimation step of estimating the cell type of the test cell based on the signal light.
2. The estimation method according to claim 1, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
3. The estimation method according to claim 2, wherein the CH ₂ stretching signal light is CH ₂ symmetrical stretching vibration signal light or CH ₂ scissor bending vibration signal light.
4. 4. The cell type of the test cell is estimated based on at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light in the signal light in the estimation step. Estimation method.
5. 5. The cell type of the test cell is estimated based on at least one of the shape and signal intensity of the second-harmonic signal light in the signal light in the estimation step. Estimation method.
6. The estimation step of estimating whether the cell type of the test cell is a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell. The estimation method described in the section.
7. 7. The estimating method according to claim 1, wherein in the estimating step, the cell type of the test cell is estimated based on the signal light and the following conditions (1) to (4):
   (1) When the signal intensity of the CH ₂ expansion/contraction signal light in the signal light is a reference value or more and the signal intensity of the second harmonic signal light in the signal light is a reference value or less, the test cell Are presumed to be pluripotent cells;
   (2) The signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or higher than the reference value, or in the signal light When the shape of the second harmonic signal light is granular, the test cell is presumed to be an ectodermal cell;
   (3) The signal intensity of the CH ₂ expansion/contraction signal light in the signal light is less than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or greater than the reference value, or in the signal light When the shape of the second-harmonic signal light is reticulated, it is presumed that the test cell is a mesodermal cell; and (4) the signal intensity of the CH ₂ stretching signal light in the signal light is a reference. When it is less than the value and the signal intensity of the signal light of the second harmonic in the signal light is less than the reference value, it is estimated that the test cell is an endoderm cell.
8. In the estimation step, the cell type of the test cell is estimated using a learned model generated by machine learning that outputs an estimation result of the cell type of the test cell based on the signal light. The estimation method according to any one of 1 to 7.
9. The estimation method according to any one of claims 1 to 8, wherein the test cell is a pluripotent cell or a differentiated cell derived from a pluripotent cell.
10. Acquisition means for acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
    A cell type estimation device comprising: an estimation unit that estimates the cell type of the test cell based on the signal light.
11. The estimation device according to claim 10, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
12. The estimation device according to claim 11, wherein the CH ₂ stretching signal light is CH ₂ symmetrical stretching vibration signal light.
13. 13. The cell type of the test cell is estimated by the estimation means based on at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light in the signal light. Estimation device.
14. 14. The cell type of the test cell is estimated by the estimation means based on at least one of the shape and signal intensity of the second-harmonic signal light in the signal light, according to any one of claims 10 to 13. The estimation device described.
15. 15. The estimating means estimates whether the cell type of the test cell is a pluripotent cell, an ectodermal cell, a mesodermal cell, or an endodermal cell. The estimation device according to one item.
16. The estimation device according to any one of claims 10 to 15, wherein the estimation means estimates the cell type of the test cell based on the signal light and the following conditions (1) to (4):
    (1) When the signal intensity of the CH ₂ expansion/contraction signal light in the signal light is a reference value or more and the signal intensity of the second harmonic signal light in the signal light is a reference value or less, the test cell Are presumed to be pluripotent cells;
    (2) The signal intensity of the CH ₂ expansion and contraction signal light in the signal light is equal to or higher than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or higher than the reference value, or in the signal light When the shape of the second harmonic signal light is granular, the test cell is presumed to be an ectodermal cell;
    (3) The signal intensity of the CH ₂ expansion/contraction signal light in the signal light is less than a reference value, and the signal intensity of the second harmonic signal light in the signal light is equal to or greater than the reference value, or in the signal light When the shape of the second-harmonic signal light is reticulated, it is presumed that the test cell is a mesodermal cell; and (4) the signal intensity of the CH ₂ stretching signal light in the signal light is a reference. When it is less than the value and the signal intensity of the signal light of the second harmonic in the signal light is less than the reference value, it is estimated that the test cell is an endoderm cell.
17. In the estimating means, the cell type of the test cell is estimated using a learned model generated by machine learning that outputs an estimation result of the cell type of the test cell based on the signal light. The estimation device according to any one of items 10 to 16.
18. The estimation device according to claim 10, wherein the test cell is a pluripotent cell or a differentiated cell derived from the pluripotent cell.
19. An observation step of observing the test cells,
    An estimation step of estimating the cell type of the test cell,
    And a selection step of selecting test cells of a predetermined cell type,
    The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
    The said estimation process is a manufacturing method of a cell implemented by the estimation method of the cell type as described in any one of Claims 1-9.
20. 20. The selection step selects test cells of the predetermined cell type by collecting test cells of the predetermined cell type or removing test cells other than the predetermined cell type. A method for producing the described cell.
21. An observation unit capable of observing test cells,
    An estimation unit for estimating the cell type of the test cell,
    Including a selection unit for selecting test cells of a predetermined cell type,
    The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
    The said estimation unit is a cell manufacturing apparatus containing the cell type estimation apparatus as described in any one of Claims 10-18.
22. The selection unit is a collection means for collecting test cells of the predetermined cell type or test cells other than the predetermined cell type, or laser irradiation capable of irradiating the test cells other than the predetermined cell type with laser. The cell manufacturing apparatus according to claim 21, which comprises a unit.
23. An observation step of observing the test cells,
    An estimation step of estimating the cell type of the test cell,
    Including a re-observation step of observing the test cells of a predetermined cell type again,
    The observing step is performed using a coherent anti-Stokes Raman scattering (CARS) microscope,
    An observation method using a coherent anti-Stokes Raman scattering (CARS) microscope, wherein the estimation step is performed by the cell type estimation method according to any one of claims 1 to 9.
24. An observation unit capable of observing test cells,
    An estimation unit for estimating the cell type of the test cell,
    A control unit capable of controlling the observation unit,
    The observation unit includes a coherent anti-Stokes Raman scattering (CARS) microscope,
    The estimation unit includes the cell type estimation device according to any one of claims 10 to 18,
    An observation apparatus using a coherent anti-Stokes Raman scattering (CARS) microscope, wherein the control unit re-observes cells estimated to be a predetermined cell type among the test cells by the observation unit.
25. An acquisition step of acquiring signal light obtained by using a coherent anti-Stokes Raman scattering (CARS) microscope of a test cell;
    A cell type including a learning step of generating a learned model that outputs an estimation result of the cell type of the test cell from the signal light, using the pair of the signal light and the cell type of the test cell as teaching data. A method of manufacturing a trained model used for estimating.
26. 26. The manufacturing method according to claim 25, wherein the signal light is at least one of CH ₂ expansion and contraction signal light and second harmonic signal light.
27. 27. The manufacturing method according to claim 26, wherein the CH ₂ stretching signal light is CH ₂ symmetric stretching vibration signal light.
28. The manufacturing method according to any one of claims 25 to 27, wherein in the learning step, at least one of the shape and signal intensity of CH ₂ expansion and contraction signal light is used as the signal light.
29. 29. The manufacturing method according to claim 25, wherein in the learning step, at least one of the shape and signal intensity of the second harmonic signal light is used as the signal light.
30. 30. The production method according to claim 25, wherein the test cells include pluripotent cells, ectodermal cells, mesodermal cells, or endodermal cells.
31. An acquisition means for acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of the test cell;
    A cell type including a learning unit that generates a learned model that outputs an estimation result of the cell type of the test cell from the signal light, using a pair of the signal light and the cell type of the test cell as teaching data. Manufacturing device for trained model used for estimation.
32. 32. The manufacturing apparatus according to claim 31, wherein the signal light is at least one of CH ₂ expansion/contraction signal light and second harmonic signal light.
33. 33. The manufacturing apparatus according to claim 32, wherein the CH ₂ expansion/contraction signal light is CH ₂ symmetrical expansion/contraction vibration signal light.
34. 34. The manufacturing apparatus according to claim 31, wherein the learning unit uses at least one of a shape and a signal intensity of CH ₂ expansion and contraction signal light as the signal light.
35. The manufacturing apparatus according to any one of claims 31 to 34, wherein the learning unit uses at least one of a shape and a signal intensity of signal light of a second harmonic as the signal light.
36. 36. The manufacturing apparatus according to claim 31, wherein the test cells include pluripotent cells, ectodermal cells, mesodermal cells, or endodermal cells.
37. An acquisition process for acquiring a signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope for the test cell;
    A program capable of executing, on a computer, an estimation process of estimating the cell type of the test cell based on the signal light.
38. An acquisition process for acquiring signal light acquired by using a coherent anti-Stokes Raman scattering (CARS) microscope of the test cell;
    On the computer, a learning process for generating a learned model that outputs an estimation result of the cell type of the test cell from the signal light, using the pair of the signal light and the cell type of the test cell as teaching data. A program that is executable.

Images