High-yield single cell screening method, system, terminal and medium

Abstract

The invention provides a high-yield single cell screening method and system, a terminal and a medium, single cells are screened based on cell images in nozzles of a printing chip after a cell solution is applied, a target cell sample is determined according to a protein yield detection result detected after the screened single cells are cultured, and a single cell recognition model is constructed; and based on the single cell identification model, determining screened high-yield single cells according to an input cell image in each nozzle of the printing chip after the cell solution is applied. According to the method, the cells in the nozzle of the printing chip are identified through the image, and the high-yield cells are screened out, so that the construction efficiency of the monoclonal cell strain can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of microfluidics and biotechnology, in particular to a high-yield single-cell screening method, system, terminal and medium. BACKGROUND

[0002] In biopharmaceuticals, target cells, such as cells with special expression, high activity, and high yield, are needed. In the cell strain construction process, the target protein expression sequence is first transfected into the host cells, and after stable genetic proliferation, single-cell sorting is performed on the transfected cells to screen high-yield single-cell clones from the cell population grown from a single cell, and the high-yield single-cell clones are expanded and banked. Among them, how to efficiently obtain high-yield single-cell clones is a key problem to improve the efficiency of the whole process, and it is necessary to purposefully select cells with high-yield characteristics for single-cell sorting to solve the problem. SUMMARY

[0003] In view of the above-mentioned shortcomings of the prior art, the purpose of the present application is to provide a high-yield single-cell screening method, system, terminal and medium to solve the technical problem of low efficiency of screening high-yield single-cell clones in the prior art.

[0004] To achieve the above-mentioned purposes and other related purposes, the present application provides a high-yield single-cell screening method, which comprises: screening single cells based on cell images in each nozzle of a printing chip after applying a cell solution, and determining target cell samples from protein yield detection results detected after culturing the screened single cells; constructing a single-cell recognition model based on each target cell sample; and determining screened high-yield single cells based on the single-cell recognition model and according to input cell images in each nozzle of the printing chip after applying the cell solution.

[0005] In an embodiment of the present application, the screening of single cells based on cell images in each nozzle of a printing chip after applying a cell solution, and the determination of target cell samples from protein yield detection results detected after culturing the screened single cells comprise: obtaining cell images in each nozzle of a printing chip after applying a cell solution collected by an imaging system as initial samples; performing single-cell indiscriminate screening according to each initial sample to obtain corresponding protein yield detection results after culturing the screened single cells; and screening a plurality of single cells as target cells based on each protein yield detection result, and marking the corresponding cell images in the nozzles of the target cells as target cell samples.

[0006] In an embodiment of the present application, the single-cell indiscriminate screening according to the cell images in each nozzle comprises: screening images containing only single cells from the cell images in each nozzle based on a single-cell indiscriminate screening algorithm, and taking the single cells corresponding to the screened images as the screened single cells.

[0007] In an embodiment of the present application, the screening of the single cells as the target cells based on the protein yield detection results and marking the cell images in the nozzles corresponding to the target cells as target cell samples comprises: sorting the protein yield detection results according to the protein yield, and screening the single cells meeting the expansion requirements as the target cells; and searching for the cell images corresponding to the target cells from the cell images in the nozzles and marking the cell images as the target cell samples.

[0008] In an embodiment of the present application, the construction of the single cell recognition model based on the target cell samples comprises: establishing a sample training set based on the target cell samples; wherein the sample training set comprises one of a first training set, a second training set and a third training set; and constructing the single cell recognition model corresponding to the sample training set; wherein the first training set is established in the following manner: marking the cell images of the cells other than the target cells in each initial sample as non-target cell samples, and obtaining the corresponding protein yield detection results; combining the target cell samples, the non-target cell samples and the protein yield detection results corresponding to the samples to construct the first training set; the second training set is constructed in the following manner: marking the cell images of the other single cells other than the target cell samples in each initial sample as non-target single cell samples, and marking the cell images of the non-single cells in the non-target single cell samples as non-single cell samples; and constructing the second training set by using the target cell samples, the non-target single cell samples and the non-single cell samples; and the third training set is constructed in the following manner: taking the target cell samples as positive samples to establish the third training set.

[0009] In an embodiment of the present application, the single cell recognition model constructed by using the first training set is used to obtain the high-yield single cell screening results and the corresponding protein yield prediction results according to the input cell images; the single cell recognition model constructed by using the second training set is used to identify whether the input cell image is a single cell, and then obtain the high-yield single cell screening results according to the cell image of the single cell; and the single cell recognition model constructed by using the third training set is used to obtain the corresponding high-yield single cell screening results according to the input cell image.

[0010] In an embodiment of the present application, an imaging system matched with the high-throughput single cell sorting print head is used to collect the cell images in the nozzles of the print chip after the cell solution is applied.

[0011] To achieve the above object and other related objects, the present application provides a high-yield single cell screening system, which comprises: a target cell determination module, configured to screen single cells based on cell images in each nozzle of a printed chip after a cell solution is applied, and determine target cell samples based on protein yield detection results detected after the screened single cells are cultured; a model construction module, connected to the target cell determination module, configured to construct a single cell recognition model based on each target cell sample; and a cell screening module, connected to the feature extraction model construction module, configured to determine screened high-yield single cells based on the single cell recognition model and according to input images of each nozzle of the printed chip after the cell solution is applied.

[0012] To achieve the above object and other related objects, the present application provides a high-yield single cell screening terminal, which comprises: one or more memories and one or more processors; the one or more memories are configured to store a computer program; and the one or more processors, connected to the memories, are configured to run the computer program to execute the high-yield single cell screening method.

[0013] To achieve the above object and other related objects, the present application provides a computer storage medium, which stores a computer program, and the computer program is configured to implement the high-yield single cell screening method when running.

[0014] As described above, the present application is a high-yield single cell screening method, system, terminal and medium, which has the following beneficial effects: the present application screens single cells based on cell images in each nozzle of a printed chip after a cell solution is applied, and determines target cell samples based on protein yield detection results detected after the screened single cells are cultured, and constructs a single cell recognition model; and the present application determines screened high-yield single cells based on the single cell recognition model and according to input cell images in each nozzle of the printed chip after the cell solution is applied. The present application screens high-yield cells by recognizing cells in nozzles of a printed chip through image recognition, thereby improving the efficiency of constructing a single clone cell strain. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A flowchart of a high-yield single cell screening method according to an embodiment of the present application is shown.

[0016] Figure 2 A structural diagram of a high-yield single cell screening system according to an embodiment of the present application is shown.

[0017] Figure 3 A structural diagram of a high-yield single cell screening terminal according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0018] The advantages and features of the present application will become apparent from the following description of embodiments of the present application, which is given for the understanding of the present application, and not for limitations of the same, and in reference to the accompanying drawings, in which: Embodiments of the present application will be described herein below with reference to the accompanying drawings. When the embodiments of the present application are implemented, other advantages and features of the present application will be clearly understood. The present application can also be embodied in different forms without departing from the spirit and essential characteristics of the present application. Embodiments of the present application can be implemented in other specific forms without departing from the spirit and essential characteristics of the present application. Embodiments described herein are illustrative of specific forms the present application might take. Many variations and modifications in form and detail can be made without departing from the spirit and essential characteristics of the present application. It is to be understood that all such variations and modifications that first appear different but that serve the same purposes as their counterparts, should be considered equivalent to those specifically described. It is therefore contemplated to cover and protect all such changes and modifications of the application, provided they come within the scope of the appended claims and their equivalents.

[0019] It is to be understood that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting. It is also possible in the present disclosure that dependent claims refer back to multiple independent claims, but such interpretation is not intended. That is, any dependent claim can be a dependent claim relating to more than one independent claim. Various modifications can be made to the embodiments and implementations of the present application without departing from the spirit and scope of the application. Therefore, other embodiments and implementations of the present application will be apparent to those skilled in the art from consideration of the specification and the practice of the application disclosed herein. It is

[0020] Throughout this specification and the claims, when it is said that a part is "connected" to another part, it includes not only the case where it is "directly connected" but also the case where it is "indirectly connected" with other elements interposed therebetween. In addition, when it is said that a certain part "includes" a certain component, it does not exclude other components unless it is specifically stated otherwise, but means that other components can also be included.

[0021] The terms first, second, third, etc. that are mentioned herein are used to explain various parts, components, regions, layers and / or sections, but are not limited thereto. These terms are used only to distinguish a certain part, component, region, layer or section from another part, component, region, layer or section. Therefore, the first part, component, region, layer or section described below can be referred to as the second part, component, region, layer or section within the scope of the present application.

[0022] Also, as used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It will be further understood that the terms "comprises", "comprising", "includes" and / or "including" when used herein, specify the presence of stated features, operations, elements, components, items, kinds and / or groups but do not preclude the presence or addition of one or more other features, operations, elements, components, items, kinds, and / or groups thereof. As used herein, the terms "or" and "and / or" are construed to be inclusive, or mean any one or any combination of the listed items. Thus, "A, B or C" or "A, B and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B and C". Exceptions to this definition apply only when the combination of elements, functions, or operations are inherently mutually exclusive.

[0023] The present application provides a high-yield single cell screening method, which screens single cells based on cell images in each nozzle of a printing chip after applying a cell solution, and determines a target cell sample according to protein yield detection results detected after culture of the screened single cells, and constructs a single cell recognition model; based on the single cell recognition model, high-yield single cells are determined according to input cell images in each nozzle of the printing chip after applying the cell solution. The present application screens high-yield cells by image recognition of cells in nozzles of a printing chip, thereby improving the efficiency of construction of a single clone cell strain.

[0024] The embodiments of the present application will be described in detail below with reference to the accompanying drawings, so that those skilled in the art in the technical field to which the present application pertains can easily implement the present application. The present application can be embodied in various different forms, and is not limited to the embodiments described herein.

[0025] As Figure 1 A flowchart of a high-yield single cell screening method in an embodiment of the present application is shown.

[0026] The method comprises:

[0027] Step S1: screen single cells based on cell images in each nozzle of a printing chip after applying a cell solution, and determine a target cell sample set according to protein yield detection results detected after culture of the screened single cells.

[0028] Wherein, the appropriate volume of the appropriate concentration of cells is added into the thermal bubble printing chip, and the high-throughput, single-cell level microfluidic design of the thermal bubble printing chip can form cells in the nozzle at one time; wherein, the thermal bubble printing nozzle uses the instantaneous high temperature of the heating film to gasify the liquid above to generate bubbles to push the liquid to flow and be sprayed from the nozzle, and then the subsequent liquid is supplemented under the action of capillary force to provide power for the continuous flow of the liquid. The thermal bubble printing nozzle is controlled by the bottom circuit. Each nozzle can form multiple cells, or only a single cell, or no cell or impurities, cell fragments, non-cell, etc. The cell images in the nozzle are collected by the imaging system, and these images will show the distribution of cells in each nozzle.

[0029] In an embodiment, step S1 comprises:

[0030] Obtaining the cell images in each nozzle of the printing chip after applying the cell solution as the initial sample, which are collected by the imaging system;

[0031] According to the cell images in each nozzle, single-cell indiscriminate screening is performed to culture the screened single cells and then detect the protein yield to obtain the corresponding protein yield detection results; single-cell indiscriminate screening can identify single cells in the nozzle and distinguish them from multiple cells, non-cell structures or fragments. This screening method aims to indiscriminately retain all single cells with different morphologies for subsequent analysis. The screened single cells are then placed in an appropriate culture environment for clonal culture. After a period of growth, protein yield detection is performed on each clone. This usually involves collecting cell samples and measuring the expression level of the target protein using specific biochemical or immunological methods. Through this step, the protein yield detection results corresponding to each clonal single cell can be obtained.

[0032] Based on the protein yield detection results, a plurality of single cells are screened as target cells, and the cell images in the corresponding nozzles are taken as target cell samples. Specifically, based on the protein yield detection results, a plurality of single cells with excellent performance (such as high protein yield) are screened as target cells.

[0033] In a specific embodiment, the high-definition single-cell images can be quickly and massively obtained by using an imaging system matched with the high-throughput single-cell sorting print head to collect the cell images in each nozzle of the print chip after the cell solution is applied. The imaging system uses a low-magnification, high-resolution, long-working-distance optical module, i.e., the imaging system uses an optical module including a high-resolution, low-magnification, long-working-distance objective lens, a barrel lens, a coaxial collimating light source, a multi-channel laser light source, a multi-channel filter, a large target camera, etc., the connection relationship and parameters of which can be adjusted as needed to realize large field of view and high resolution imaging, matched with the high-throughput print chip to complete high-throughput single-cell sorting and result identification.

[0034] In a specific embodiment, the single-cell indiscriminate screening based on the cell images in each nozzle includes:

[0035] Based on the single-cell indiscriminate screening algorithm, the images containing only single cells are screened from the cell images in each nozzle, and the single cells corresponding to the screened images are taken as screened single cells. This algorithm only eliminates non-single-cell images, and retains all morphological single cells indiscriminately. Using this method, the most extensive single-cell samples can be obtained to the greatest extent.

[0036] In a specific embodiment, the way of detecting the protein yield of the screened single cells to obtain the corresponding protein yield detection results includes: placing the screened single cells in a 96-well plate for clonal culture, detecting the protein yield of the clonal strains in each well after a certain period of growth, and determining the target cells meeting the amplification requirements according to the detection results.

[0037] In a specific embodiment, the way of screening a plurality of single cells as target cells based on the protein yield detection results of each protein yield detection result and taking the cell images in the corresponding nozzles as target cell samples includes:

[0038] The protein yield detection results of each protein yield detection result are sorted according to the protein yield, and the single cells meeting the amplification requirements are screened as target cells; the cell images in the nozzles corresponding to the target cells are found from the cell images in each nozzle and marked as target cell samples. It should be noted that the standard of the protein yield detection results of the single cells meeting the amplification requirements is determined according to the requirements.

[0039] Step S2: constructing a single-cell recognition model based on the target cell sample set.

[0040] In an embodiment, step S2:

[0041] establishing a sample training set based on the target cell samples; wherein the sample training set includes one of: a first training set, a second training set, and a third training set.

[0042] The single cell recognition model is constructed by using the sample training set.

[0043] The process of constructing the single cell recognition model based on the target cell sample involves three different training set construction methods. Each method aims to train the model by different sample combinations to more accurately identify the target cell.

[0044] The first training set is established by:

[0045] Since the screened single cells are cultured for protein yield detection, and based on the protein yield detection results of each protein yield detection, a plurality of single cells are screened as target cells, the cell images of each single cell remaining after screening the target cells are marked as non-target cell samples, and the corresponding protein yield detection results are obtained as part of the training set; In addition, the cell images of other non-single cells in each initial sample are also marked as non-target cell samples, and the corresponding cells can also be subjected to protein yield detection to obtain protein yield detection results, or the protein yield results are set to a specific value (e.g. 0 or a negative value).

[0046] The first training set is constructed by combining each target cell sample, each non-target cell sample, and the protein yield detection results corresponding to each sample.

[0047] The second training set is established by:

[0048] The cell images of other single cells screened from each initial sample except the target cell samples are marked as non-target single cell samples, and the cell images of non-single cells in them are marked as non-single cell samples.

[0049] The second training set is constructed by combining each target cell sample (representing single cells and high-yield single cells), non-target single cell sample (representing single cells but not high-yield single cells), and non-single cell sample (representing non-single cells).

[0050] The third training set is constructed by:

[0051] The third training set is established by using each target cell sample as a positive sample.

[0052] Step S3: Based on the single cell recognition model, determine the screened high-yield single cells according to the input cell images in each nozzle of the printed chip after applying the cell solution.

[0053] In an embodiment, the single cell recognition model constructed by the above three different training sets can achieve specific functions.

[0054] The single cell recognition model constructed by the first training set is trained by using each target cell sample and each non-target cell sample, and the protein yield detection results corresponding to each sample.

[0055] The single-cell recognition model considers whether the cell is a target cell (i.e., whether it is a high-yield cell) and the corresponding protein yield. It receives a cell image as input and outputs two results:

[0056] 1. Classification result of whether it is a high-yield single cell (e.g., represented by a binary classification label, such as "high yield" or "non-high yield").

[0057] 2. Predicted protein yield value.

[0058] The single-cell recognition model constructed using the second training set is trained using target cell samples, non-target single cell samples, and non-single cell samples.

[0059] The single-cell recognition model is first used to identify whether the input cell image is a single cell, and then further processes the images identified as single cells to screen high-yield single cells. The model receives a cell image as input and outputs a binary classification result indicating whether the image represents a single cell. After identifying single cells, further judgment is made on these single cell images to determine whether they are high-yield cells.

[0060] The single-cell recognition model constructed using the third training set is trained using target cell samples. The single-cell recognition model obtains the corresponding high-yield single cell screening result based on the input cell image.

[0061] To better describe the high-yield single cell screening method, the following specific embodiments are combined for description.

[0062] Embodiment: CHO cell monoclonal sorting. After the host cells are transfected with the target protein plasmid and cultured and passaged to stabilize the activity, a high-throughput single cell screening instrument with an imaging system with high resolution and a large field of view is used to sort the transfected cells into a microplate. During the sorting process, high-definition images of each sorted single cell can be obtained. The imaging device is used to confirm the monoclonality, and after a certain period of culture, the protein yield of the single clone strains in the microplate is determined. According to the detection results, the target single clone strains that meet the requirements of the downstream amplification production process are selected, labeled, and the sorting pictures of these single clone strains are collected. These picture data are used as a sample library to obtain a cell recognition model with high activity and high yield. Using this model for single cell sorting, the cell population can be screened in advance during the sorting stage, increasing the number of target cells that can be obtained from each single clone sorting, thereby improving the efficiency of the overall production process.

[0063] Similar to the principles of the above embodiments, the present application provides a high-yield single cell screening system.

[0064] The following specific embodiments are provided in conjunction with the accompanying drawings:

[0065] like Figure 2 A schematic diagram of the structure of a high-yield single-cell screening system according to an embodiment of the present invention is shown.

[0066] The system includes:

[0067] The target cell identification module 1 is used to screen single cells based on cell images in each nozzle of the printed chip after the cell solution is applied, and to identify target cell samples based on the protein yield detection results of the screened single cells after culture.

[0068] Model building module 2 is connected to the target cell determination module 1 and is used to build a single-cell recognition model based on each target cell sample;

[0069] Cell screening module 3, connected to the feature extraction model construction module 2, is used to determine high-yield single cells to be screened based on the single-cell recognition model and the images of each nozzle of the printed chip after the cell solution is applied.

[0070] It should be noted that, as should be understood Figure 2 The division of modules in the system embodiment is merely a logical functional division. In actual implementation, they can be fully or partially integrated into a single physical entity, or they can be physically separated. Furthermore, these units can be implemented entirely in software through processing element calls; they can be implemented entirely in hardware; or some units can be implemented by processing element calls to software, while others are implemented in hardware.

[0071] Since the implementation principle of this high-yield single-cell screening system has been described in the foregoing embodiments, it will not be repeated here.

[0072] In one embodiment, the step of screening single cells based on cell images within each nozzle of the printed chip after the application of cell solution, and determining target cell samples based on protein yield detection results of the screened single cells after culture, includes: acquiring cell images within each nozzle of the printed chip after the application of cell solution, collected by an imaging system, as initial samples; performing non-discriminatory screening of single cells based on each initial sample, and then culturing the screened single cells to obtain corresponding protein yield detection results; screening multiple single cells as target cells based on each protein yield detection result, and marking the cell images within the corresponding nozzles as target cell samples.

[0073] In one embodiment, the single-cell indiscriminate screening based on the cell images in each nozzle includes: based on a single-cell indiscriminate screening algorithm, screening images containing only single cells from the cell images in each nozzle, and using the single cells corresponding to the screened images as the screened single cells.

[0074] In an embodiment, the screening of the single cells as the target cells based on the protein yield detection results and marking the cell images in the nozzles corresponding to the target cells as target cell samples comprises: sorting the protein yield detection results according to the protein yield, and screening the single cells meeting the expansion requirement as the target cells; and searching for the cell images corresponding to the target cells from the cell images in the nozzles and marking the cell images as the target cell samples.

[0075] In an embodiment, the constructing of the single cell recognition model based on the target cell samples comprises: establishing a sample training set based on the target cell samples; wherein the sample training set comprises one of a first training set, a second training set and a third training set; and constructing the single cell recognition model corresponding to the sample training set; wherein the first training set is established in the following manner: marking the cell images of the cells other than the target cells in each initial sample as non-target cell samples, and obtaining the protein yield detection results corresponding to the non-target cell samples; combining the target cell samples, the non-target cell samples and the protein yield detection results corresponding to the samples to construct the first training set; the second training set is constructed in the following manner: marking the cell images of the other single cells other than the target cell samples in each initial sample as non-target single cell samples, and marking the cell images of the non-single cells in the non-target single cell samples as non-single cell samples; and constructing the second training set by using the target cell samples, the non-target single cell samples and the non-single cell samples; and the third training set is constructed in the following manner: taking the target cell samples as positive samples to establish the third training set.

[0076] In an embodiment, the single cell recognition model constructed by using the first training set is used to obtain the high-yield single cell screening results and the corresponding protein yield prediction results according to the input cell images; the single cell recognition model constructed by using the second training set is used to identify whether the input cell image is a single cell, and then obtain the high-yield single cell screening results according to the cell image of the single cell; and the single cell recognition model constructed by using the third training set is used to obtain the corresponding high-yield single cell screening results according to the input cell image.

[0077] In an embodiment, an imaging system matched with the high-throughput single cell sorting print head is used to collect the cell images in the nozzles of the print chip after the cell solution is applied.

[0078] The high-yield single cell screening method provided by the embodiments of the present application can be implemented on the terminal side or the server side. As for the hardware structure of the high-yield single cell screening terminal, please refer to Figure 3An optional hardware structure diagram of the high-yield single-cell screening terminal 1000 provided by the embodiment of the present application is shown in FIG. 1. The terminal 1000 can be a mobile phone, a computer device, a tablet device, a personal digital processing device, a factory background processing device, etc. The terminal 1000 includes at least one processor 1001, a memory 1002, at least one network interface 1009, and a user interface 10010. The various components in the apparatus are coupled together by a bus system 1005. It can be understood that the bus system 1005 is used to realize the connection communication between the components. The bus system 1005 includes a data bus, a power bus, a control bus, and a status signal bus, but for the purpose of clear illustration, all the buses are marked as the bus system in FIG. 1. Figure 3

[0079] The user interface 1009 can include a display, a keyboard, a mouse, a trackball, a click gun, a key, a button, a touchpad, or a touch screen, etc.

[0080] It can be understood that the memory 1002 can be a volatile memory or a non-volatile memory, or both. The non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), which is used as an external cache. By way of example but not limitation, many forms of RAM can be used, such as static random access memory (SRAM), synchronous static random access memory (SSRAM). The memory described in the embodiment of the present application is intended to include but not limited to these and any other suitable categories of memory.

[0081] The memory 1002 in the embodiment of the present application is used to store various categories of data to support the operation of the terminal 1000. Examples of these data include any executable programs for operating on the terminal 1000, such as an operating system 10021 and an application program 10022. The operating system 10021 contains various system programs, such as a framework layer, a core library layer, a driver layer, etc., for realizing various basic services and processing hardware-based tasks. The application program 10022 can contain various application programs, such as a media player (MediaPlayer), a browser (Browser), etc., for realizing various application services. The high-yield single-cell screening method provided by the embodiment of the present application can be included in the application program 10022.

[0082] ​The method disclosed by the embodiments of the present application can be applied to the processor 1001 or implemented by the processor 1001. The processor 1001 can be an integrated circuit chip having a signal processing capability. In the implementation process, each step of the above method can be completed by an integrated logic circuit or an instruction in the form of software in the processor 1001. The processor 1001 can be a general processor, a digital signal processor (DSP), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, etc. The processor 1001 can implement or execute the disclosed methods, steps and logic block diagrams in the embodiments of the present application. The general processor 1001 can be a microprocessor or any conventional processor, etc. In combination with the steps of the accessory optimization method provided by the embodiments of the present application, the hardware decoding processor can be directly embodied to complete the execution, or the hardware and software modules in the decoding processor can be combined to complete the execution. The software module can be located in a storage medium, which is located in a memory. The processor reads the information in the memory and combines the hardware to complete the steps of the foregoing method.

[0083] In the exemplary embodiments, the terminal 1000 can be one or more application specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), or the like for executing the foregoing method.

[0084] Those of ordinary skill in the art can understand that all or part of the steps of the foregoing method embodiments can be completed by computer program related hardware. The foregoing computer program can be stored in a computer readable storage medium. When the program is executed, the steps of the foregoing method embodiments are executed; and the foregoing storage medium includes various storage media that can store program codes, such as ROM, RAM, magnetic disk or optical disk, etc.

[0085] In the embodiments provided in the present application, the computer readable and writable storage medium can include read-only memory, random access memory, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage device, flash memory, U disk, mobile hard disk, or any other medium capable of storing desired program code in the form of instructions or data structures and capable of being accessed by a computer. In addition, any connection can be appropriately referred to as a computer readable medium. For example, if instructions are sent from a website, server or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL) or wireless technology such as infrared, radio and microwave, the coaxial cable, fiber optic cable, twisted pair, DSL or wireless technology such as infrared, radio and microwave is included in the definition of the medium. However, it should be understood that the computer readable and writable storage medium and the data storage medium do not include connections, carriers, signals or other temporary media, but are intended for non-transitory, tangible storage media. As used in the application, magnetic disks and optical disks include compact disks (CD), laser disks, optical disks, digital versatile disks (DVD), floppy disks and Blu-ray disks, wherein magnetic disks usually magnetically copy data, and optical disks use lasers to optically copy data.

[0086] The present application has the following advantages:

[0087] 1. The present application uses an optical module with low magnification, high resolution and long working distance to realize large field of view and high resolution imaging, which matches high-throughput printing chips to complete high-throughput single cell sorting and result identification.

[0088] 2. The present application clonally cultures the sorted single cells, and after a certain period of culture, high-yield cells are screened according to cell output, and then the sample library is optimized using the single cell sorting picture corresponding to the high-yield cells, and the AI recognition standard is re-established.

[0089] 3. The present application uses a combination of traditional algorithms and deep learning to identify high-yield single cells, so that the sorted cells have high activity.

[0090] To sum up, the high-yield single cell screening method, system, terminal and medium of the application, by screening single cells based on the cell image in each nozzle of the printing chip after applying the cell solution, and determining the target cell sample by the protein yield detection result of the screened single cells after culture, a single cell recognition model is constructed; based on the single cell recognition model, the high-yield single cells screened are determined according to the input cell image in each nozzle of the printing chip after applying the cell solution. The application screens the high-yield cells by image recognition of the cells in the printing chip nozzle, so as to improve the efficiency of the construction of the single clone cell strain. Therefore, the application effectively overcomes the various shortcomings in the prior art and has a high industrial utilization value.

[0091] The above embodiments only exemplarily illustrate the principles and effects of the application, and are not used to limit the application. Any person skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the application. Therefore, all equivalent modifications or changes completed by those skilled in the art without departing from the spirit and technical thought disclosed by the application should be covered by the claims of the application.

Claims

1. A method for screening high-yield single cells, characterized in that, The method includes: Single cells are screened based on cell images within each nozzle of the printed chip after the cell solution is applied, and the target cell sample is determined by the protein yield detection results of the screened single cells after culture. A single-cell identification model was constructed based on each target cell sample; Based on the single-cell recognition model, high-yield single cells are selected by analyzing the cell images within each nozzle of the printed chip after the application of cell solution.

2. The high-yield single-cell screening method according to claim 1, characterized in that, The process of screening single cells based on cell images within each nozzle of the printed chip after applying cell solution, and determining the target cell sample based on protein yield detection results after culturing the screened single cells, includes: The images of cells within each nozzle of the printed chip after the application of the cell solution, acquired by the imaging system, are used as initial samples. Single-cell non-discriminatory screening was performed on each initial sample. The selected single cells were then cultured and protein yield was measured to obtain the corresponding protein yield results. Based on the protein yield detection results, multiple single cells were selected as target cells, and the cell images in the corresponding nozzles were labeled as target cell samples.

3. The high-yield single-cell screening method according to claim 2, characterized in that, The single-cell indiscriminate screening based on cell images within each nozzle includes: Based on the single-cell indiscriminate screening algorithm, images containing only single cells are selected from the cell images in each nozzle, and the single cells corresponding to the selected images are used as the selected single cells.

4. The high-yield single-cell screening method according to claim 2, characterized in that, The step of selecting multiple single cells as target cells based on the protein yield detection results and labeling the corresponding cell images within the nozzle as target cell samples includes: The protein yield test results were sorted according to the protein yield size, and single cells that met the amplification requirements were selected as target cells. Find the cell image corresponding to the target cell from the cell images in each nozzle and mark it as the target cell sample.

5. The high-yield single-cell screening method according to claim 2, characterized in that, The single-cell identification model constructed based on the target cell sample includes: A sample training set is established based on the target cell sample; wherein, the sample training set includes: a first training set, a second training set, and one of a third training set; A single-cell recognition model is constructed using the corresponding sample training set; The methods for establishing the first training set include: Cell images in each initial sample other than the target cell sample are labeled as non-target cell samples, and the corresponding protein yield detection results are obtained; The first training set was constructed by combining the protein yield detection results of each target cell sample, each non-target cell sample, and each sample. The methods for constructing the second training set include: In each initial sample, cell images of other single cells besides the target cell sample are labeled as non-target single cell samples, and cell images of non-single cells are labeled as non-single cell samples. A second training set was constructed using each target cell sample, non-target single cell sample, and non-single cell sample. The methods for constructing the third training set include: The third training set is established by using each target cell sample as a positive sample.

6. The high-yield single-cell screening method according to claim 5, characterized in that, The single-cell recognition model constructed using the first training set is used to obtain high-yield single-cell screening results and corresponding protein yield prediction results based on the input cell images. The single-cell recognition model constructed using the second training set is used to identify whether the input cell image is a single cell, and then to obtain high-yield single-cell screening results based on the single-cell cell image. The single-cell recognition model constructed using the third training set is used to obtain the corresponding high-yield single-cell screening results based on the input cell image.

7. The high-yield single-cell screening method according to claim 2, characterized in that, Images of cells within each nozzle of the printed chip are acquired using an imaging system matched to the high-throughput single-cell sorting printhead after the application of the cell solution.

8. A high-yield single-cell screening system, characterized in that, The system includes: The target cell identification module is used to screen single cells based on cell images in each nozzle of the printed chip after the cell solution is applied, and to identify target cell samples based on the protein yield detection results of the screened single cells after culture. The model building module is connected to the target cell determination module and is used to build a single-cell recognition model based on each target cell sample. The cell screening module, connected to the feature extraction model construction module, is used to determine high-yield single cells to be screened based on the single-cell recognition model and the images of each nozzle of the printed chip after the cell solution is applied.

9. A high-yield single-cell screening terminal, characterized in that, include: One or more memories and one or more processors; The one or more memories are used to store computer programs; The one or more processors, connected to the memory, are used to run the computer program to perform the high-yield single-cell screening method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The device contains a computer program that, when executed, implements the high-yield single-cell screening method as described in any one of claims 1 to 7.

Images