Automated collection of specified number of cells

Abstract

To provide an automated method and system to isolate and collect cells using computerized analysis of images of cells and their surroundings obtained from a digital imaging device or system.SOLUTION: A "microwell array," which can comprise an elastomeric grid-like indentations or "wells", are used. Many, most, or all of the wells in a microwell array can contain a releasable, microfabricated element, which is referred to as a "raft." Embodiments of the disclosed subject matter provide a system and method for cell collection that includes computerized identification and collection of rafts with isolated single cells or one or more specific groups of cells, eliminating the need for continuous human identification and selection.SELECTED DRAWING: Figure 1B

Description

[Technical field]

[0001] (Related Applications) This application is a continuation of U.S. Provisional Patent Application No. 62 / 416,773, filed November 3, 2016. No. 6,349,633, the disclosure of which is incorporated herein by reference in its entirety. , which claims the benefit of U.S. Provisional Patent Application No. 62 / 416,775, filed November 3, 2016. This application claims the benefit of US Provisional Patent Application No. 2016 / 013696, the disclosure of which is incorporated herein by reference in its entirety. This application claims the benefit of U.S. Provisional Patent Application No. 62 / 430,094, filed December 5, 2003, which is hereby incorporated by reference in its entirety. The disclosure of which is incorporated herein by reference in its entirety. This application was filed on June 28, 2017. This application claims the benefit of U.S. Provisional Patent Application No. 62 / 526,177, filed on the same day, the entire disclosure of which is incorporated herein by reference in its entirety. The bodies are incorporated herein by reference.

[0002] (Statement Regarding Federally Funded Research) This invention was made in part without funding from the National Institutes of Health, Grant No. 2R42GM106421. This invention was made with Government support under the terms of the present application. The Government has certain rights in this invention. [Background technology]

[0003] Selection and isolation of multiple single cells from a mixed population is a common procedure carried out throughout biomedical research. For example, genetically engineered cells Propagation of cell lines, stem cell-derived cell lines, or cell lines propagated from patient cell lines. During development, multiple single cells are isolated and then cloned to form a homogenous population. Fluorescence-activated cell sorting (FACS), limiting dilution, panning ), column chromatography and magnetic separation to separate non-adherent Various strategies exist for selectively identifying and collecting cells; New techniques based on electrochemical and dielectrophoretic techniques show promise in this field. To address the need to collect pure or enriched populations of rent cells, The researchers either dissociate the cells from their growth surface to create a cell suspension, or Unfortunately, enzymatic or mechanical release is difficult to achieve. This leads to loss of cell morphology, removal of cell surface markers, damage to the cell membrane, changes in cell physiology and survival. Imposing serious drawbacks, including loss of abilities.

[0004] To address this issue, we developed a lab-based ... Polydimethylsiloxane (PDMS) containing releasable, microfabricated elements called rafts. Cell Separation and Recovery System Using Poly(PDMS) Microwell Arrays One version of such a system is the CELLRAFT™ system. Rafts are suitable growth regions for multiple single cells or large colonies. The size of PDMs can vary from tens to hundreds of microns to provide a range of rafts. The microwell array can be visualized, for example, using an inverted microscope. A number of individual rafts containing desired cells can be isolated by various methods, such as mechanical compression or continuous shaking. By applying a small amount of energy to the microwell array, the actuators can be made to react with each other upon mechanical distortion of the microwell array. In one example, multiple individual rafts containing multiple desired cells can be released from the rafts. or visualized by a technician and then placed under a microscope objective. The attached probes are released from the array by mechanically pushing them into the raft. It is possible. Summary of the Invention [Means for solving the problem]

[0005] An embodiment of the present invention provides images of cells and their surroundings obtained from a digital imaging device or system. An automated method for isolating and collecting cells using computer analysis of images of the surrounding tissue. Embodiments of the present invention provide a method and system for forming a formed elastomeric casing. The present invention utilizes "microwell arrays" which can contain small depressions or "wells". Many, most, or all of the wells in the microwell array are releasable and microfabricated. The present invention may include a raft of elements that are spaced apart from each other. The disclosed embodiments include a rough sample having a plurality of isolated single cells or one or more specific cell groups. The present invention provides a system and method for cell collection, including computerized identification and collection of cells. This provides a single set of options, eliminating the need for continuous human identification and selection.

[0006] In some examples, the system comprises: The system is configured to acquire an image of the microwell array. and a method for releasing cell rafts from a microwell array. and a computer having at least one processor and a memory. and a computer system for detecting the presence of the microwell array using an imaging device. acquiring one or more images of the microwell array; and analyzing one or more images of the microwell array. The selected cell rafts were identified and extracted from the microwell array by analyzing the The actuator is programmed to release the selected cell raft. There are.

[0007] The system may be adapted to move the imaging device or the microwell array, or both. and one or more motors configured to move one or more of the microwell arrays. To obtain more images, the microwell array must be divided into multiple fields of view. The field was designed to include multiple cell rafts collectively, and each field was and controlling one or more motors to direct the imaging device to a field of view of the microwell array. By controlling the imaging device to capture each image in the field of view, the array scan is The method may include performing a scan.

[0008] The system may include a microscope objective for the imaging system. Acquiring one or more images of the array of fields of view includes at least one Autofocus a microscope objective using one or more focus positions from adjacent fields Autofocus of the microscope objective may include scanning multiple sample focal positions. Sampling multiple sample images at different positions and calculating the focus scores of the sample images This may include interpolating the current focus position from the focus score.

[0009] Identifying the selected cell rafts involves subdivision of the cell rafts for each cell raft. Counting the number of cells depicted in the image and determining which at least one contains a single isolated cell. The sub-images may include identifying a single cell raft. Counting the number of pixels is done by applying an intensity threshold to the sub-images to produce a binary image. creating a binary image of a unique object; identifying the unique object in the binary image; Identifying single-cell rafts involves counting the number of objects. A confidence score was determined that indicates the confidence in the decision that the sample contained a single isolated cell. This may include determining

[0010] Identifying the selected cell rafts involves subdivision of the cell rafts for each cell raft. detecting a marker depicted in the image; and gating the cell raft based on the detection of the marker. Identifying the selected cell rafts can include gating the selected cell rafts. and assigning the selected cell rafts to mapped locations on a collection plate. The system can include a mechanical cell raft collector. The computer system then identifies the selected cell rafts after their release from the microwell array. Control the mechanical cell raft collector to collect and plate the selected cell rafts. Controlling a mechanical cell raft collector to deposit cells at mapped locations It can be configured as follows.

[0011] In some examples, the system is adapted to monitor a cellular process. For example, the system may be configured to administer a drug or reagent, or to induce cytotoxicity, cell proliferation, or the like. The time course of a cell or group of cells for changes in gene expression or cell state, such as proliferation or metabolism. The system can be configured to perform quantitative analysis of the microwell array. and an imaging device configured to capture the microwells. The system comprises at least one processor, a processor for processing the at least one cell raft within the microwells of the array. The computer system includes a processor and a memory. a first time, an imaging device was used to confirm that at least one cell raft was involved in an intracellular process. The computer is configured to determine that the cell or cells in the sample contain one or more cells undergoing a The system comprises one or more cellular rafts in which at least one cell raft is undergoing an intracellular process. In response to determining that the at least one microwell array contains a cell on the The location of the cell rafts is then stored in the computer at least one time after the initial time. The data system uses the location to identify the location of at least one cell raft. and using an imaging device to monitor intracellular processes at least It is configured to acquire one or more images of a single cell raft.

[0012] Monitoring intracellular processes may include storing one or more images and detecting one or more Each of the further images is recorded with the capture time and the The intracellular process can include associating the location of at least one cellular raft. The monitoring includes displaying at least one cell in a graphical user interface. This can include presenting one or more images of the raft and the initial image.

[0013] At least one cell raft contains one or more cells undergoing an intracellular process. The determination of the presence of a cellular raft is made by illuminating at least one cell raft with light of a particular wavelength; and detecting, using an imaging device, at least one cell raft in response to illuminating at least one cell raft. Obtain initial images of cell rafts; Monitoring the intracellular process may include detecting a group signature. In the graphical user interface, the fluorescence intensity of one or more images can be calculated. It may include presenting a graph over time.

[0014] At least one cell raft contains one or more cells undergoing an intracellular process. Determining that at least one cell raft contains only a single cell means that the cell raft is a single-cell raft that contains only a single cell. and quantifying baseline fluorescence levels using the initial images. Monitoring the intracellular process may include determining whether the one or more images are a match between the initial image and the intracellular process. determining a difference in the measured fluorescence of

[0015] In at least some embodiments, a computer-implemented method for harvesting cells is provided. The proposed method records information about microwell arrays containing multiple rafts. and identifying one or more cellular rafts from among multiple rafts. The system as described herein includes a method for collecting single cell rafts. A single cell raft can be used to identify a single cell that associates with Rafts with cells that have been isolated, and often these are the cell rafts that will be selected. However, systems according to embodiments of the present invention may also, or instead, Rafts that have a certain number of cells other than one cell, or, as another example, a certain type of single cell and other types of cells or combinations of specific numbers of cells of different types. The system then controls and uses the actuators to separate the raft. We release cell rafts from the microwell array by using a magnet as an example. The system can then collect cell rafts using a 3D filter before moving on to the next raft. The release of the microwell array can be confirmed at any time. Associated with or connected to a processor that executes a computer-implemented method The information can be stored in a storage device.

[0016] In some embodiments, the storage of information about the microwell array comprises storing information about the microwell array in a Identification and memorization of the size of the microwell array, and the optimal focus position of the microwell array Identification and storage of the optimal exposure and sectioning of the microwell array sectioning, and matching a set of microwells to the field of view. In some embodiments, the microprocessor includes a memory for storing the translation required for the Storing information about the well array includes calculating and storing offsets for the actuators. Includes.

[0017] In some embodiments, the identification of cell rafts is performed by quantification of multiple labels of a microwell array. Segmenting rafts and dividing multiple rafts into multiple rafts In some embodiments, counting comprises counting the cell nuclei relative to a fluorescence threshold. Performing a watershed transform to define the nucleus In some embodiments, the identification of released cell rafts involves detection of a marker. Thus, further including gating or determining which cellular rafts are released. The markers are then used to measure the intensity of one or more color channels (which , which may show fluorescence caused by staining) and / or the size of the cell nucleus, the cell itself The term may include the body or non-nuclear structures within a cell.

[0018] Workstation, personal computer, handheld computer, or embedded A computer system, such as an embedded processing system, functions as part of a system. and / or imaging devices, systems and / or hardware, actuators and other and when executed by a processor, A non-transient computer that causes a system to execute all or part of a process according to an embodiment Contains or has access to program code. The code can be stored in a storage device. [Brief description of the drawings]

[0019] [Figure 1A] FIG. 1 illustrates an exemplary system for harvesting cells. [Figure 1B] FIG. 1 illustrates an exemplary system for harvesting cells. [Figure 1C] FIG. 1 illustrates an exemplary system for harvesting cells. [Figure 2A] FIG. 1 is an isometric view of an exemplary apparatus for harvesting cells. [Figure 2B] FIG. 1 is an isometric view of an exemplary apparatus for harvesting cells. [Figure 3A] FIG. 1 shows an exemplary method for harvesting cells. [Figure 3B] FIG. 1 shows an exemplary method for harvesting cells. [Figure 4A] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Figure 4A] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Figure 4B] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Figure 4C] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Figure 4D] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Figure 4E] FIG. 2 is a diagram illustrating a workflow showing the interaction between user inputs, graphical user interface functions, hardware, and software for a system according to an exemplary embodiment of the present invention. [Diagram 5] FIG. 1 shows a flow diagram of an exemplary method for determining array location and type of a microwell array. [Figure 6] FIG. 1 shows a flow diagram of an exemplary method for navigating a microwell array. [Figure 7A] FIG. 1 shows an exemplary method for autofocusing a microscope lens onto the field of view of a microwell array. [Figure 7B] FIG. 1 shows an exemplary method for autofocusing a microscope lens onto the field of view of a microwell array. [Figure 7C] FIG. 1 shows an exemplary method for autofocusing a microscope lens onto the field of view of a microwell array. [Figure 8] FIG. 1 shows a flow diagram of an example method for identifying single-cell rafts in a microwell array. [Figure 9] FIG. 13 shows screenshots of exemplary screens of a graphical user interface for selecting cell rafts and mapping collection plates. [Figure 10] FIG. 11 is a screenshot of another exemplary screen of a graphical user interface. [Figure 11A] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11B] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11C] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11D] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11E] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11F] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11G] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11H] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. [Figure 11I] FIG. 11A shows a block diagram and a series of images (FIGS. 11B to 11I) illustrating an experimental example. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will now be more fully described with reference to the accompanying drawings, in which embodiments of the invention are shown. However, the present invention may be embodied in many different forms and should not be construed as limited to the specific embodiments set forth herein. The present invention should not be construed as being limited to the embodiments set forth herein. are for purposes of illustrating particular embodiments only and are not intended to limit the invention. As used herein, the singular forms "a," "an," and "the" The terms "a," "b," and "c" are intended to include the plural forms unless the context clearly indicates otherwise. As used herein, the term "comprises" or "comprising" means any of the features, steps, , action, element, or component, but not one or more other features, steps, does not preclude the presence or addition of other elements, operations, components, or groups thereof. Comparative terms such as "above," "below," "less than," "more than," and quantitative terms are equivalent. Therefore, "less" is intended to encompass the concept of In this case, it means not only "less than" but also "less than or equal to."

[0021] Unless otherwise defined, all terms used herein (including technical and scientific terms) The terms "compound" and "compound expression" have the same meaning as commonly understood by a person skilled in the art to which the present invention pertains. Furthermore, terms used herein shall have the same meaning as those used herein within the context of this specification and the related art. shall be construed to have a meaning consistent with those of the terms used herein and not expressly defined herein. It will be understood that the terms are not to be construed in an idealized or overly formal sense unless and until the elements are When something is referred to as being "connected" or "coupled" to another element, it refers to something that is directly connected to the other element. It will also be understood that the elements may be connected or joined, or that intervening elements may be present.

[0022] Single cell isolation allows for molecular analysis and characterization of individual cells (e.g., DNA, RNA, proteins). Colonies were generated by clonal expansion from individual cells, including cells after transfection. Each well ("microwell array") can be useful for a wide variety of scientific studies, including the acquisition of Microwell arrays containing multiple cell rafts in a single well are useful for isolating and collecting single cells. It has been shown that microwells can be sampled using standard microscopic manual methods. Isolation of single cells by separation and collection from the array allows for the production of large numbers of single cells for analysis or expansion. It may take a significant amount of time to obtain cells. Details of microwell arrays with cell rafts that can be used with the systems and methods No. 9,068,155, incorporated herein by reference. An automated system for collecting cells from microwell arrays seeded with cells of interest is Reduce the time and effort required to perform scientific and medical research with cells while It also includes additional features such as cross-channel imaging.

[0023] A system for collecting cells according to an exemplary embodiment of the present invention includes an actuator; an imaging device; and a processor connected to the actuator and the imaging device; As shown in, for example, FIGS. 3A-3B, 4A-4E, 5, 6, 7A and 8, Executable computer program code stored therein for carrying out the computer-implemented method. The method and system of the present invention can be operated using a microwell array. Identify single isolated cells present on individual cell rafts in the laminae and identify single cell laminae. As an example, the cells are collected by collecting the suction tubes. It has been done.

[0024] 1A-1C are diagrams illustrating an exemplary system 100 for cell collection. The system 100 may include a number of cells, cell colonies, and rough specimens having various numbers of different types of cells. FIG 1A is an overview of a system 100. The system 100 includes a computer system 102, an equipment assembly 104, and an experimental environment 10 6 (e.g., one or more pieces of laboratory equipment, such as a power supply and environmental control system), and the user The instrument assembly 104 includes a microwell array 112. An optional adapter plate and a microwell array 112 are selected and released. and a collection plate 114 for receiving the collected cell rafts. The collection plate 114 comprises: The samples may be organized into a standardized format, such as an SBS collection plate. However, the system 100 may instead be configured to use any suitable sample, such as a PCR strip tube. Any suitable collection structure may be used.

[0025] Typically, a user 108 deposits or cultures cells in a medium on a microwell array 112. The cells are then seeded onto the microwell array 112 and allowed to settle into individual cell rafts. is mounted in an adapter plate 110 of the system 100 for scanning and image analysis. The system 100 then controls and uses the actuators to 3 to release the cell rafts from the microwell array 112 and remove the cell rafts using, for example, a magnet. For example, the actuator may be a needle. one or more needles or similar devices configured to move the raft to release it In some examples, the system 100 may include a magnetic wand. ic wand) and a separate actuator for moving the stage. Multiple actuators, including actuators, imaging optics, and other mechanical parts of the system. Includes a actuator.

[0026] FIG. 1B is a block diagram of computer system 102. 02 includes at least one processor 120, a memory 122, and a A controller 124 that is executed as a computer program using the A graphical user interface (GUI) 126 is included in the The system 102 may be a desktop computer equipped with a monitor, keyboard and mouse. Alternatively, the computer system 102 may be a laptop or tablet computer. The computer system 102 may be a computer, a mobile device, or any other suitable device. For example, a universal serial bus (USB) cable may be used to connect the device assembly 104 is operatively coupled to

[0027] The controller 124 acquires one or more images of the microwell array 112. , and by analyzing the images, the selected cell rafts in the microwell array 112 are identified. to control the instrument assembly 104 to identify and release selected cell rafts. The controller 124 is programmed as shown in FIGS. The method examples further described with reference to Figures 7A and 8 can be implemented. 26 presents various control and result screens to the user 108 and receives input from the user 108. is configured to receive

[0028] FIG. 1C is a block diagram of the instrument assembly 104. The instrument assembly 104 includes a microphone Individual cell rafts 130 on the low-well array 130 are imaged and placed in a collection plate 114. Various methods for selectively releasing cell rafts 130 from the microwell array for placement in the microwell array are described. For example, the equipment assembly 104 may include a power breakout board (po wer breakout board) 138, and various Control boards (e.g., PS3 control board 132, PS3 XYZ control board 134, and PS3 F The motor control board may include a motor control board 136. T that allows the various components of the instrument assembly 104 to be controlled or addressed May include TL and shutter functions.

[0029] The equipment assembly 104 includes a digital camera 140 or other suitable imaging device, a communications hub ( For example, a USB hub 142, a fluorescence LED engine 144, and a light The fluorescent LED engine 144 may include a light guide adapter 146. A plurality of nanotubes configured to illuminate the microwell array 112 via a photodiode 146. -band LEDs may be included.

[0030] The instrument assembly 104 includes a motorized actuator configured to move the microscope objective 152. A microscope system (e.g., , internal inverted digital microscope). Typically includes a camera 140 and a fluorescent LED engine 144 and the microscope system are arranged in an epi-fluorescence configuration. In some examples, the microscope system may detect cell rafts 130 from the microwell array 112. The release probes 154 are configured to individually release the The focus motor 150 can be operated by the fast focus motor 150.

[0031] In some instances, the microscope system may achieve 2 microns per pixel for a given field of view. A microstructure with a raft of 200 microns by 200 microns in size, with a resolution of 100 nm or less. At least about 4×4 array features on the well array, and rafts of 100 μm Supports imaging of areas with a typical 8 x 8 array when using a 100 x 100 micron array The microscope system also uses bright-field imaging (i.e., white light illumination and white light emission). and acquiring images in one or more fluorescence emission channels. In some cases, exposure times across all channels can be supported. Assuming a time interval of 750 ms, the instrument assembly 104 will Scanning the entire microwell array in 20 minutes for both chromatographic and bright field. can be done.

[0032] The release probe 154 typically releases an acid when exposed to saline or cell culture medium. In some examples, the release probe 154 is made of stainless steel 10 mm thick. The release probe 154 is a 0 micron needle. Thus, for example, it may have a possible travel distance of at least 15 mm in the X and Y directions.

[0033] The equipment assembly 104 is used to move the gantry assembly. a belt drive 156 for imaging the microwell array 112; and a bright field light for illuminating the microwell array 112 during imaging. Move the LED 158 and the magnetic wand 162 to collect the cell rafts after release. The gantry assembly includes a linear actuator 160 configured to The entry assembly may use a lead screw instead of a belt drive, or any other suitable motor. The linear actuator 160 can be, for example, a microwell actuator. The magnetic wand 162 is configured to move in and out of the ray 112 and collection plate 104. The stepping motor may be a stepping motor.

[0034] The instrument assembly 104 is placed under the collection plate 114 and receives the magnetic flux from the magnetic wand 162. The collection magnet 164 collects the cell rafts on the collection plate 114. The magnet 164 repels the magnet in the magnetic wand 162, causing the cell rafts to adhere to the surface of the collection plate 104. It can have a polarity opposite to that of the magnetic wand 162 to attract it to the bottom. The gas wand 162 is typically used with released cell rafts or collection plates 104. The medium can be sterile (e.g., while being removed from the instrument) to avoid contaminating the medium being transferred. Materials for the magnetic wand 162 Also, generally, contact with the culture medium results in a detectable decrease in cell viability or proliferation, or A detectable decrease in the performance of molecular biology reagents such as Taq polymerase or reverse transcriptase The dose is selected so as not to cause a decrease in

[0035] 2A and 2B are isometric views of an exemplary device for collecting cells. 2A is an exemplary implementation of the equipment assembly 104 shown in FIG. The mounting plate 110 and the collection plate 114 are disposed on a horizontal XY stage 148. Some components, such as electronic control boards 132, 134, and 136, control the XY stage. The XY stage 148 is disposed under the microwell array 112. and the collecting plate 104. The XY stage 148 is configured to move the Positioning the cell raft for imaging (aligning the cell raft with the microscope objective 152) , releasing the cellular raft (aligning the cellular raft with the release probe 154), and Place the magnetic wand 162 and the collection plate 104 on the collection magnet 164. The position of the sensor can be electronically controlled.

[0036] The gantry assembly, including the belt drive 156, is mounted vertically above the XY stage 148. The gantry assembly positions the bright field LED 158 for imaging. The Gantt 162 is also configured to move laterally to position the magnetic wand 162 for sensing the magnetic field. The assembly is magnetically attached onto a microwell array 112 to collect the rafts during release. The gas wand 162 is then positioned over the collection plate 114. The magnetic wand 162 is positioned to deposit cell rafts at selected locations on the collection plate 114. Ku.

[0037] The camera 140 and the autofocus motor 150 are, for example, 50 can move vertically relative to the XY stage 148. The fluorescent LED engine 144 and the liquid light guide port are arranged. The deport 146 is disposed under the XY stage 148 and includes a fluorescent filter cube 170. The fluorescent filter cube 170 is coupled to, for example, the fluorescent LED engine 144. The light is then allowed to reach the microwell array 112 and is then captured by the camera. The fluoroscopic imaging system is configured to prevent radiation from reaching 140 nm.

[0038] FIG. 2B shows a cutaway view 172 of the microscope objective 152 and the release needle 154. As shown, the release probe 154 is positioned so that it does not intersect with the field of view of the microscope objective lens 152. The release probe 154 is offset from the optical axis of the microscope objective lens 152. Without intersecting the field of view, the user can visualize the release of cell rafts in real time. As a result, the system uses imaging to confirm release, and the It may be necessary to move the microwell array 112. First, to improve imaging speed, the release probe 1 is placed outside the field of view of the microscope objective lens 152. It may be useful to place 54.

[0039] In some other examples, the release probe 154 is positioned within the field of view of the microscope objective lens 152. By placing a release probe 154 within the field of view of the microscope objective lens 152, This allows the user to visualize the release of cell rafts in real time; A suitable arrangement may require imaging through an acrylic window, which may reduce the amount of excitation and emission light. This can reduce transmission and require longer integration times during the scan.

[0040] In either case, the center of the field of view of the microscope objective 152 and the microwell array 11 It may be useful to calibrate the offset between the puncture position of the release probe 154 on the Calibration should be performed, for example, after every needle change, or at the beginning of every experiment, or after every production In some examples, the controller 124 of FIG. is programmed to carry out the

[0041] For example, the controller 124 may move the microwell array 112 to The field of view of lens 152 is positioned at the array border and the microscope objective lens is aligned. The array boundary can then be pierced with a release probe 154. Thereafter, the controller 124 positions the puncture position within the field of view of the microscope objective lens 152. 158 to move the microwell array 112 to ) an image is acquired and the image is analyzed, for example by segmenting the image, to locate the puncture location. The controller 124 can then calculate the offset. In the example, the controller 124 may be configured to move to a different location a specified number of times. Repeat the process and add the offset position, e.g., by averaging the offset positions. Based on this, a calibration distance is determined.

[0042] 3A-3B illustrate an exemplary method for harvesting cells. FIG. 3 is a flow diagram of an exemplary method 300 for cell collection from a low well array. 1, and the method 300 may be implemented by the controller 124 of FIG. This will be described with respect to the system 100.

[0043] The method 300 includes initializing the system (302) and setting up the experiment (304). These are further described below with reference to FIGS. 4A-4B. For example, the array scan (306) may be initiated by prompting the user 108 using the GUI 126. ) is performed. If array scanning is not performed, method 3 is performed. 00 receives a user command (308), for example, using the GUI 126. This includes navigating the microwell array 112 by user commands. The navigation is performed when the user navigates the microwell array 112. This can include presenting an image of the microwell 122 array in real time. The method 300 then proceeds to select a method for collection of cells, which is selected, for example, using the GUI 126. The method includes receiving a user selection of one or more cell rafts (310).

[0044] When performing an array scan, the method 300 scans the microwell array without user input. 3. Performing an array scan by navigating the array 112 (314). The method 300 then comprises: detecting one or more cell roughs using image analysis for cell collection. FIG. 3B illustrates the use of image analysis to select cellular rafts (316). 3 is a flowchart of an exemplary method 350 for detecting single-cell rafts. (352), gating of single cell rafts based on marker detection (354), and The labeled single-cell rafts are then mapped to their location on the collection plate (356). Gating of single-cell rafts is generally based on the detection of fluorescent markers. The method further comprises selecting a subset of the detected single-cell rafts.

[0045] Array scanning, single cell detection, and automated or semi-automated cell raft selection are shown in Figure 4. C-4D. In some instances, array scanning and The identification of cellular rafts is performed in parallel. For example, the controller 124 may run multiple threads to It may be performed in parallel or may use any suitable parallel processing technique.

[0046] The method 300 includes releasing and transferring (312) the selected cell rafts. This is further described below with reference to FIG. 4E. The method 300 may, for example, The user input, captured images, the location of the detected single cell rafts, and the released cells Export data describing the experiment, such as a raft record, and any other appropriate data. The export of data may optionally include copying the Storing data in a file or transmitting data to a remote system for storage This may include:

[0047] As can be seen in FIGS. 3A-3B, a system according to an exemplary embodiment can Real-time imaging, which may include storing the images in a storage device ("real-time imaging mode"). ) can be used in manual selection mode where the user identifies single cell rafts during It is also used in a computer-implemented method for collecting cells, called the "Tree Image Analysis" mode. According to some embodiments, the user 108 may control the system via the GUI 126. The system 100 is operated in a cytometry image analysis mode or a real-time imaging mode. In the real-time imaging mode, the user 108 selects the microwell array 11 2. Selectively navigate the imaging field of view within the In cytometry image analysis mode, you can select rafts to separate them. A full scan of the cross-well array is performed, followed by quantitative image processing (e.g., of the fluorescent images). The process facilitates user selection of cells for isolation or automatic selection of cells for collection. .

[0048] In cytometric image analysis mode, the system scans the microwell array and storing information about the microwell array including the plurality of rafts in a storage device; The image is analyzed using the information obtained, and one or more cells are identified from among the multiple rafts. The present invention relates to a method for identifying a cell raft having a microfluidic device, the method comprising: In some embodiments, the cell rafts are released from the microwell array. Storing the relevant information includes calculating and storing offsets for the actuators. In some embodiments, the system relies on the known positions of particular rafts to predict future To facilitate future manipulations, each of the multiple microwells in the microwell array may be The system determines an address for, for example, a fluorescent color or Brightfield channel intensity, size of nuclei or other structures, or a combination of these markers Based on the detection of position-based markers, the actuator-driven release of cell rafts is detected. In some embodiments, the system can perform a magnet-based The cell rafts containing the magnetic nanoparticles were collected and then arbitrarily confirmed that the cell rafts were indeed released. It can be confirmed.

[0049] 4A-4E are diagrams illustrating user input, graphics, and other aspects of a system in accordance with an exemplary embodiment of the present invention. Physical user interface functions, hardware, and software interactions 1 is a workflow diagram illustrating the example system 100 of FIG. Each workflow diagram is divided horizontally into four regions. The top region is the user The area below the top area shows the GUI 126. The next area shows the operations performed using hardware (e.g., equipment assembly 10 4) The bottom area shows the operations performed by the controller 124. Indicates the action taken.

[0050] FIG. 4A shows the start of an example experiment. A user 108, for example, removes cells from an incubator and The user 108 places the microwell array 112 in front of the plate and releases the cells from the plate. The user 108 then assembles the instrument and seeds the cells onto the microwell array 112. It provides power to the bridge 104 and computer system 102, and the controller 124 and G Starts UI126. During initialization, GUI126 displays the initialization splash screen ( tialization splash screen, while the instrument assembly 104 can display a The controller sets the parameters to move the motor to the home position and then to the load position. 124 is an equipment assembly 104 including a motor controller, LED lighting, and a camera; Initiate and verify communication.

[0051] FIG. 4B shows the next step in the example experiment: setting up the experiment. Prompt user 108 to place well array 112 on the microscope stage. User 1 08, the microwell array 112 is mounted on the XY stage 148 (e.g., the mounting plate 110 The GUI 126 can be used in either cytometry or real-time analysis mode. Prompt the user 108 to select one, receive a selection from the user 108, and then to display a screen indicating that the system is calibrating the array position of the microwell array 112. Display.

[0052] The controller 124 controls the instrument assembly to acquire an image of the microwell array 112. 104 and to identify the orientation and type of the microwell array 112. and executing an array location and type routine, including: For example, characterizing the number and size of cell rafts on the microwell array 112 The array position and type routine can perform image segmentation on, for example, a brightfield image. Perform autofocus and navigation routines by running the cell rafts An exemplary array location and type routine can be seen in FIG. In some examples, the system 100 may include a microphone within a field of view. Visual control of the brightfield image and stage motors for positioning the corners of the Lowell array 112 Click on it to control feedback and identify array location and type The user 108 is provided with a user interface control for:

[0053] If the user 108 selects the real-time imaging mode, the controller 124 A timed image acquisition routine is executed. The user 108 sets the imaging parameters and field of view. , the image is captured in real time by the instrument assembly 104 , which captures the microwell array 112 . Inspect the image when navigating. Controller 124 operates the viewing position. In response to user input from the GUI 126, the XY stage 148 and the autofocus mode The controller 124 controls the autofocus and navigation router 150. 112 to navigate the microwell array and visualize the cell roughness in a new perspective. After inspecting the images, the user 108 selects one or more If the user 108 selects the cytometry imaging mode, The GUI 126 allows the user 108 to set imaging parameters for the array scan. Optionally present a screen for the acquisition of the image. The acquisition parameters are instead read from a file. The digital signal may be transmitted directly to the server or may be received over a network connection.

[0054] FIG. 4C illustrates a user 108 selecting a cytometry mode, array scanning, and single cell detection. The GUI 126 shows the next step in the example experiment when you select the output. The controller 124 displays a status screen on the display screen. For example, the controller 124 does not overlook cell rafts. Maximize the number of cell rafts in each field while ensuring that The array area is divided into separate fields of view by making the number of pixels smaller.

[0055] In general, the controller 124 maximizes the number of cell rafts in each field of view and selects the next set of cells. Optimize the raft's position within each field of view to calculate the movement required to image the raft. To achieve this, iterative state positioning and raft segmentation are performed. The controller 124 may, for example, determine the cell raft dimensions and the spacing between the cells in the array. alignment with the septum; nominal margin between all cell rafts in the field and the edge of the field; state motor based on the accuracy of the state motor movement; and the computation time to detect cell rafts in the images. Based on this, iterative state localization and raft segmentation can be performed. Instead, the controller 124 may select a visual field that ensures an overlap of at least one cell raft width. The cells can move a given distance between the fields and resolve overlapping cell raft images during processing ( can be resolved.

[0056] The instrument assembly 104 moves the XY stage 148 to the first field of view. The laser 124 can include autofocus and imaging, which can include identifying cell rafts within the field of view. and a navigation routine, and then the controller 124 executes an image acquisition routine. The image acquisition routine checks, for example, whether a particular fluorescent channel is specified. Depending on which imaging parameters one or more specific light sources are turned on, and then, for example, The method may include controlling the camera 140 to be specified for a fixed exposure time.

[0057] The controller 124 then controls the XY stage 148 to move to the next field of view and The focus and navigation routine and the image acquisition routine are then repeated. The scan may be performed, for example, on the entire microwell array 112 or on specific portions of the microwell array 112. The image continues until a certain end condition is reached, such as until a certain part is captured or until a time limit is reached. Array scanning, for example, processes a row in one direction, then the next row in the opposite direction. During the array scan, the controller The controller 124 assigns an address to each detected cell raft. The address is useful and As a result, for example, images and other data of cell rafts can be associated with cell rafts (as ) and position cell rafts for release and migration. Cut.

[0058] Once the scan is complete, the GUI 126 prompts the user 108 for single cell detection parameters. The user 108 may select, for example, color channels and The thresholds for the color channels can be input. The controller 124 controls a single Perform image analysis of the images obtained during the scan to identify cell rafts. For example, The controller 124 uses the extracted sub-images of the selected image to generate a pixel value within each cell raft. Typically, the cell nuclei of the microwell array 112 can be segmented. An image of a field may contain multiple cell rafts. It is possible to extract sub-images of specific cell rafts. This involves segmenting the image to identify the location of cell rafts and then identifying specific and isolating portions of the image that represent single cell rafts. is discussed further below with reference to FIG. 8. The controller 124 then Metrics for segmented cell rafts and cell nuclei in selected images (me trics) can be calculated and compiled.

[0059] FIG. 4D shows how the user 108 selects the cytometry mode, marker gating, and plate mark gating. The next step in the example experiment is shown below when marker gating is selected. 6 presents a screen to the user 108 for specifying the marker gating parameters. The user 108 defines a fluorescence gate. For example, the user 108 specifies a threshold color intensity. The controller 124 may be configured to determine the number of gated single cell rafts. The user calculates and edits metrics for each parameter, and the GUI 126 presents a screen that displays the results.

[0060] Markers are characteristics of cells, which may be nucleic acids, proteins, or other types of organic or inorganic compounds. These are represented by cellular characteristics such as morphological features and the size and structure of organelles. These proteins will be present in a variety of cells, and their detection within a given cell will identify or classify the given cell. ify), or to other products that may not contain certain markers or may contain them to a different extent. The presence, relative amount or quantitative amount of a marker is used to distinguish a cell type from another cell type. Detection can be achieved using a variety of methods and materials. These include detecting cells, organelles, or the cytoskeleton, nucleus, mitochondria, and various other organelle compartments. Dyes, including fluorescent dyes that label other cellular features such as the organellar compartment Such dyes are well known in the art. The optically detectable target sequence may be represented by a nucleic acid having a sequence complementary to the target sequence. Possible probes, antibodies against specific marker proteins, or fluorescent motifs (e.g. green fluorescent proteins), antigens (e.g. HisX6) or enzyme activities (e.g. luciferase or The gene expressing the marker is a non-native part, such as alkaline phosphatase. A transgenic approach designed to contain a native molecule The present invention may include the use of methods and materials for detecting a target molecule.

[0061] In the present context, a marker may also refer to one or more chemical or parenteral molecules alone. The fluorescent dye may be in combination with a complementary moiety, which determines whether cells are present. or whether a given cell is dead, alive, intact, not intact, or the state or process of other cells. The term marker may also be used to determine whether a fluorescent or other optical The term may be used herein to describe a signal, such as an optically detectable signal, This is detected by image analysis in an automated system and used to identify or classify a given cell. or to distinguish it from other types of cells or from cells having different characteristics or properties. is used to.

[0062] During plate mapping, the GUI 126 maps cell rafts to wells within the collection plate 114. For example, the user 108 may select a graphical representation of a cell raft ( The cell rafts were manually selected from the graphical display and placed in wells within the collection plate 114. In another example, the controller 124 may assign a cell raft to the graphical representation. The cell rafts can be randomly assigned to available wells in the collection plate 114. The controller 124 may, for example, determine whether both are single-cell rafts (e.g., at least one have a certain level of confidence) and a particular marker is present (e.g., By selecting each cell raft in which the detected marker is present at least at a threshold level, This allows the selection of cell rafts based on user-specified criteria. The La 124 compiles a list of cell raft locations for the collection plate wells.

[0063] In some instances, the GUI 126 will display all the segments within the currently selected marker gate. For cell rafts, the approximate location of the cell rafts on the microwell array 112 and one or more is a two-dimensional histogram or scattergram of the fluorescence intensity in two imaging channels. Display the distribution diagram. Each cell raft within the gated population is identified by a clear visual indicator (e.g., a single It is represented on the scatter plot by a single yellow filled circle, and its (x, y) position is The fluorescence intensity of the cell raft in each imaging channel corresponding to the two axes is equal to the fluorescence intensity. is integrated over the entire cell raft and represents all single cells for that imaging channel. Scaling against maximum integrated cell raft strength within the resident population of Other indicators along each axis, such as colored flags, are scaled by the user. allows easy switching of imaging channels for a given axis.

[0064] In some examples, the GUI 126 allows the user 108 to place marker gates on the fluorescence intensity scatter plot. A user interface control may be provided that allows the user to create Quadrant - A single left click on the scatter plot defines the position of a set of crosshairs. Double-clicking on any of the four defined quadrants will create a new gate in that quadrant. Each quadrant can be used to create a separate gate. Line - Left-click twice on a scatter plot to create an infinite line across the scatter plot. Define the trajectory. Double-clicking on either side of the line creates a new gate. can be used to create separate gates. Ellipse - First left click on the scatter plot defines the center of the ellipse. Second left click The click defines one axis and radius of the ellipse relative to the center. The third mouse click , defines the orthogonal radius relative to the first axis. Left-click twice inside the ellipse to open a new Given the methodology for defining the ellipse, the ellipse may extend beyond the bounds of the scatter plot. There is a saying. Polygon - Left-click and hold on the scatter plot to define the vertices of a polygon. Close the polygon To do this, left-click on the original vertex. The edges of a polygon cannot cross each other, but they can be There is no limit to the number of vertices you can define. Left-click twice inside a polygon to create a new gate. Create.

[0065] In some examples, the system 100 allows the user 108 to plot multiple images on a single two-dimensional scatter plot. It allows to define ellipses and polygons. If such gates overlap, When you select an area where the two gates overlap, a new gate is created based on the original gate geometry. In some cases, a previously created marker gate (parent gate) is created as the current marker gate. When selected as a focus gate, the GUI 126 will display the cellular rafts within the selected gate. The new marker gate placement is then displayed in the scatter plot and array plot. When drawn on a gate and used to create a new marker gate (child gate), Additional cell rafts within the endogenous population of all single cells may fall within the range of the child gate configuration. Despite the fact that there is a marker gate, only the cellular rafts within the parent marker gate are included in the child gate. are screened to be

[0066] FIG. 4E shows the next step in the example experiment: raft release and translocation. The beads are released from the microwell array 112 and mapped onto a collection plate 114. The equipment assembly 104 is moved to a desired position under the control of the controller 124. align the release probe 154 with the initial calibrated cell raft location and place the magnetic wand 162 The instrument assembly 104 is then aligned with the release probe 154. is activated to release the cell rafts and the magnetic wand 162 is moved into the microwell array 112. The wand is then lowered and then raised away from the microwell array 112.

[0067] The controller 124 executes a cell raft release and collection verification routine, for example: After releasing the cell rafts, the controller 124 may then control the cell rafts to image the released cell rafts. The controller 124 can then move the XY stage 148 to obtain the Image analysis was performed on the captured images to determine whether cell rafts were depicted in the acquired images. If the cell raft is not depicted, the controller 124 can determine whether the cell raft is depicted. When cell rafts are depicted, the controller 124 can, for example, For example, a release probe 154 may be used to wait a specified number of times before stopping to report an error. The release routine can then be repeated.

[0068] After confirming the release, the controller 124 transfers the released cell rafts to the collection plate 11. The cells are then transferred to the wells assigned to that cell raft in the 4. Update the status screen to show which cell rafts have been released and which remain to be released. The controller 124 then selects another cell raft for release. Align the release probe 154, activate the release probe 154, and then release the cell roughness. The magnetic wand 162 is controlled to collect the signal.

[0069] The controller 124 repeats the process for each cell raft selected for release. Once each cell raft selected for release has been moved, the GUI 126 indicates completion. The user 108 can then select the microwell array 112 and collection profile from the system. Extract the rate.

[0070] The user 108 may store a summary file, for example on a local hard drive. The summary file may be stored in any suitable format, such as a GUI 126. Compatible with the database structure, typically: 1) an imaging channel used for cell identification; ;2) the imaging channel used for cytometric analysis; 3) the cytometry channel Description of the fluorescence signal gate of the selected marker used for the diameter); 4) the exposure time used for each imaging channel; 5) the total scan time (if performed 6) a microwell array 112 assigned to each well of a collection plate 114 7) the location of single cell rafts within each imaging channel for each collected cell raft 8) the average fluorescent signal intensity (relative fluorescent units - RFU) in the channel; and 9) the The labeling process performed when the microwell array 112 was approved for placement on the system The controller 124 searches and collects For each single cell raft captured, all active images selected by the user 108 are Export the image from the image channel in a standard image file format compatible with the image. You can sport it.

[0071] FIG. 5 illustrates an exemplary method for determining array location and type of microwell array. 5 is a flow chart of a method 500 performed by the controller 124 of FIG. The method 500 may be implemented in a variety of ways, including, for example, in a system 100 of FIG. For example, known array geometry, manufacturing tolerances, and cell rafts detected in the images were used. Searching for specific locations (e.g., corners) of the microwell array 112 in a dynamic manner Includes.

[0072] The method 500 includes positioning the XY stage 148 at an initial position (502), e.g., at the microwell alignment position. The method includes moving one of the corners of the ray 112 to an estimated position for imaging. is responsible for the autofocus of the microscope system, the execution of image acquisition routines, and the acquisition of images. Segmenting the acquired image to determine the location of cell rafts within the image (504). The method 500 includes calculating the average fineness, which is, for example, the side length (L) in the case of a square cell raft. The method can include calculating vesicle raft size.

[0073] The method 500 may, for example, determine whether the acquired image has corners. 2. Move the XY stage 148 to find a corner of the microwell array 112. For example, determining whether one or more cellular rafts are If it is detected within a threshold distance (e.g., distance 2*L) of the top of the image, the method 500 moves to the top. This can include determining to move (e.g., by a distance L) to the left of the image. If one or more cell rafts are detected within the threshold distance, the method 500 may The method may include determining to move leftward (a distance L).

[0074] When moving the XY stage 148, the method may be performed to move the microwell array 112 closer to a corner. and moving the XY stage 148 so as to form a microwell array (508). 112. Once a corner is found, the method 500 continues by searching for the type of array. To determine the position and orientation, the XY stage 148 is arbitrarily moved to a final position based on the position of the corner. The method includes moving the camera to a final position, autofocusing at the final position, and capturing an image 510.

[0075] The method 500 determines the type and orientation of the array based on the image at the final position. For example, the type of array may include a particular corner ( For example, by encoding a pattern in the upper left corner of the microwell array 112 Cell rafts can be encoded as either present or absent (i.e. Provides digital information on whether a microwell is filled or not The method 500 further comprises analyzing additional regions of the microwell 112 to determine type and orientation. For example, the method 500 may include determining whether to image the microorganism (514). This can be repeated for each of the four corners of the well array 112.

[0076] FIG. 6 illustrates an example method 600 for an autofocus and navigation routine. 6 is a flowchart illustrating a method 600 performed by the controller 124 of FIG. and will be described with respect to the system 100 of FIG.

[0077] The method 600 includes autofocusing a particular field of view of the microwell array 112. and acquiring an initial image (602). The method 600 includes detecting cell rafts in the initial image. The method 600 includes detecting (604) a number of cell rafts in a field of view that maximizes the number of cell rafts in the field of view. For example, the cell rafts may be first If the image is oblique, moving the XY stage 148 will change the field of view. This may include moving the XY stage 148 to center the cell raft at the boundary. The method 600 includes acquiring 608 a new image after the movement. The method 600 may, for example, include detecting cellular rafts in the new image (610). By assigning an address to each cell raft detected in the old image, The method further includes indexing (612) the cell rafts. The addresses can be, for example, For example, it may be a row address and a column address.

[0078] For example, while imaging the microwell array 112, each increment in the field of view The following process can be performed with (increment): 1. The autofocus routine runs and selects the best autofocus in the stack of images. Brightfield images with Cass scores are selected for segmentation. 2. Using OpenCV routines, use the Otsu method to extract the 3 Two pixel intensity thresholds that maximize the interclass variance across the two classes of pixels Only the upper threshold is applied to convert the bright-field grayscale image to a binary image. will be done. 3. Use OpenCV routines to find each uniquely detected object, i.e. A bounding box for a set of adjacent bright pixels is calculated. 4. The bounding box is subsequently band-pass filtered to determine its area, which is the nominal cell raft surface. Any objects that deviate from the nominal volume by more than 15% are removed; and the cell rafts are nominally Being a top square, objects whose X and Y dimensions deviate from each other by more than 10% are also eliminated. 5. Determine the tilt of the array in the field of view by selecting the array that contains the most cell raft objects. A row of cell rafts containing the cell raft is selected. A linear regression to the center of the cell raft's bounding box is used to A first order polynomial is calculated, and the gradient defines the gradient direction. 6. The four corner cell raft objects are selected and are then grouped into cell rafts based on their inclination direction. Their bounding boxes are used to calculate the margins between the edges of the field of view. 7. The XY stage is moved to center the nominal cell raft subarray within the field of view. The best focus point is then interpolated during the autofocus routine. ) the Z focus drive is adjusted. 8. A new brightfield image is acquired and steps 2-4 are repeated. 9. Segmented cell rafts are sorted by rows and columns and are visualized by comparing the previous field of view and XY stage movement. The data is appropriately indexed based on the index assigned to the direction of movement.

[0079] FIG. 7A illustrates a flow diagram of an exemplary method 700 for autofocusing a microscope objective. 7 is a flow chart of a method 700 that may be performed by the controller 124 of FIG. This can be done and will be described with respect to the system 100 of FIG.

[0080] In some examples, the system 100 performs a rough autofocus function at the beginning of an experiment. Perform fine autofocus with any shift in the field of view of the imaging microscope afterwards. The first autofocus procedure for a given experimental run is This is performed after the well array 112 is inserted and the system begins initialization. The configuration file maps the microscope objective and system field of view to a single reservoir. The upper left quadrant of a microwell array, or the upper left reservoir of a four reservoir microwell array. The X and Y positions of the XY stage 148 that are aligned with the camera can be specified. Navigate to the location and perform the following rough autofocus steps:

[0081] 1. The focus drive is the focus that is defined in the configuration file. Move to the minimum position of the search range. 2. Brightfield images are taken using, for example, 2x binning and dynamically determined exposure times. It is obtained. 3. In parallel: a. Image data is transferred from the camera and a pixel is calculated based on the variance of pixel intensities across the image. A focus score is calculated that quantifies the contrast of the image. b. The focus driver is raised, for example, by 100 μm. 4. The focus driver has reached the upper limit of the search range defined in the configuration file. If not, repeat steps 2 and 3. If so, the software will Extract the focus position using the score.

[0082] System 100 then focuses the microscope objective lens onto the particular field of view using any suitable technique. For example, the controller 124 can autofocus the lens on the Z axis. You can run a Z-stack routine to find the position where the image is in focus. The stack routine uses a fixed distance between the lower and upper search limits along the Z axis and the sample focus position. In some examples, the array scan During scanning, search limits may be applied to all fields of view within the microwell array 112. The Z-stack routine then acquires images and calculates a focus score at each sample focus position. Then, the best focus score is used to extract the position or the sample focus score is calculated. Other examples of search routines include the Golden Finder. These include the golden search method and Brent's method.

[0083] FIG. 7 illustrates a method 700 for Z-stack imaging to leverage information from adjacent fields of view. The method 700 may be considered a modification to the routine, for example, during an array scan. It may be useful, so the search limits could potentially be narrowed to improve autofocus speed or accuracy or can improve both.

[0084] The method 700 includes moving the XY stage 148 to a new field of view and adjusting the microscope objective lens. Move the lens along the Z axis to the end of the possible focus range (e.g., bottom or top). In some examples, the method 700 may include: Based on the focus position from at least one adjacent field of view of the microwell array 112, Determine the possible focus range (or otherwise determine some sample focus positions) )

[0085] For example, focus positions from adjacent fields are typically not available at the beginning of an array scan. However, it is usually available as the scan progresses. In some cases, method 7 00 is the Z axis based on the average of arbitrary focus positions already determined within eight adjacent fields of view. Select the upper and lower search limits along , as well as the step size between sample focus positions. Figure 7B shows the range of sample focus positions along the Z axis.

[0086] The method 700 includes acquiring an image at a sample focal position using the camera 140. (704). The method 700 includes performing a microscopy with a fixed step size, e.g., 28 microns. By moving the mirror objective upwards, the microscope objective is moved to the next sub-axis along the Z axis. The method 700 includes moving the image data to a sample focus position (706). 7. Transferring (708) the image to the computer system 102 and calculating a focus score for the image. Any suitable type of quantitative focus metric may be used as the focus score. For example, the focus score can be determined based on the gray level variance. can be determined.

[0087] The method 700 may also include moving the microscope objective to the next sample focus position in parallel. Optionally, the method 700 may include calculating a focus score. If necessary, the method 700 may include waiting for both operations to be completed (712). The objective is to determine whether the objective is at the top (or bottom) of its possible focal range. , determining (714) whether there are any further sampling focus positions to sample; If there are more sampling focus positions to sample, the method 700 The imaging and focus score calculations are then repeated for each position.

[0088] The method 700 uses the Z position of the sample focus and the focus score at the sample focus. 7C shows the Z-axis location of best focus (on the horizontal axis) using the An example plot of sample focus positions (on the vertical axis) and focus score (on the vertical axis) is shown. For example, Then, the method 700 calculates the best focus score (Z1, FS1 in FIG. 7C) and the sample focus. Two sample foci on either side (along the Z axis) of the point (Z2, FS2 and Z3, FS in Figure 7C) 3) to extract the Z-axis position of the sample focus. Then, the interpolated The Z-axis position (Z focus ) is calculated as follows:

number

[0089] The SIGN operator generates a positive or negative sign depending on the difference between the Z axis positions of Z2 and Z3, and Zstep is is the step size between sample focus positions. The interpolated focus positions are then calculated using the Any suitable imaging for 100 may be used.

[0090] Although system 100 has been described as autofocusing at each field of view, In general, system 100 may use any suitable technique for focusing a microscope objective. For example, the system can be used to measure (x,y) positions in the array. At any array location in the array, by fitting a surface to a subset of the focal planes A focal plane interpolation can be performed. Surface fitting can be done, for example, to fit the physical Based on a model derived from the properties (dimensions, elasticity, etc.) and mass of the fluid in the reservoir For example, the system 100 can use thin plate splines, which can be expressed as follows: The least squares minimization between the surface and the control points represents the minimum bending (i.e., the minimum second derivative) to calculate a smooth surface; between the control points, the radial basis function U(r) = r 2 Using ln(r); Use LU decomposition to compute the second order polynomial weights and first order coefficients of the surface as a function of (x, y) .

[0091] FIG. 8 illustrates an exemplary method 80 for identifying single-cell rafts in a microwell array. 8 is a flowchart of a method 800 performed by the controller 124 of FIG. This can be done and will be described with respect to the system 100 of FIG.

[0092] The method 800 includes extracting, for a selected cell raft, a sub-image of the cell raft. (802). The method 800 includes removing (e.g., by masking) an outer boundary of the image. and applying an intensity threshold (e.g., a user-specified threshold) to the image to generate a binary image (804). Creating an image (806).

[0093] The method 800 may include, for example, detecting a binary image using any suitable object detection algorithm. The method 800 may further include, for example, identifying a unique object within the image (808). Identify unique objects that have fewer pixels (e.g., 10 pixels) Filter out unique objects identified based on size by discarding them. The method 800 includes filtering (810) the remaining objects after filtering. Counting the cells (812) and, if only a single cell is shown, the selected cell label. This includes identifying the

[0094] If the selected cell raft is identified as a single cell raft, the method 800 determine a confidence score indicating the confidence of the determination that the identified cell raft is a single-cell raft; For example, to assign a confidence level to the classification of a single cell. Selected cell rafts were then further characterized using their fluorescence morphology. For each candidate single-cell cell raft: Local maxima in the grayscale pattern of detected fluorescent objects are smoothed and These are identified using a combination of dilation processes. A 9x9 flat structuring element (SE) has zero defined at the origin. A grayscale dilation is applied to the masked image (I) using SE.

number

[0095] In some instances, any cell raft candidates remaining after the culling process The complements are classified according to their confidence level and the marker gating Filter based on confidence, size, or maximum intensity before including in a preliminary single cell population for It is available for tagging.

[0096] In general, the controller 124 may use any suitable image analysis to identify single cell rafts. In some examples, the controller 124 may be configured to determine whether or not the cell rafts are present. For a sub-image of, we can threshold that sub-image and find the number of discrete objects If it is one after thresholding, calculate the distance transform of the complement of the threshold image to produce the negative distance transform and set the background pixels to -Inf (or other appropriate placeholder value). identifying and marking local minima to distinguish single cells from clusters of cells; and A watershed transformation is performed to draw a boundary around the cells.

[0097] FIG. 9 illustrates a cell imaging system for a real-time imaging mode, such as that described above with reference to FIG. Graphical user interface for selecting rafts and mapping collection plates FIG. 9 is a screenshot of an exemplary screen of the system control and sample capture. An exemplary “imaging” image (e.g., from GUI 126 of FIG. 1) combining multiple elements of FIG. 1 shows a GUI screen for explaining the user interface for adjusting the focus of the imaging optics or It not only allows the autofocus to be initiated, but also allows the user to move the field of view in each direction. or move it to any position with a single mouse click. The user interface further provides interface controls. Turn on or off the red, green, and blue fluorescent channels and adjust the exposure time and It allows the user to set the intensity, gain, and binning for each fluorescence channel. It allows for independent adjustment of exposure time for each active imaging channel.

[0098] As shown in FIG. 9, the user can click or select a raft of interest in any image. You can then select a raft, inspect the zoomed image of the raft, and select a specific raft for isolation. The user interface displays the location of the currently displayed field of view on the microwell array. The GUI screen also shows the collection plate (a 96-well format is shown as an example) The system displays a map of the single-cell rafts selected for isolation, and the user can select the single-cell rafts by This allows for assignment to a specific location within the collection plate identified at the beginning of the experiment.

[0099] Using display elements from Figure 9, the GUI also provides a The GUI screen may generate a screen that can be used to select the rafts to be separated. The system displays the results and allows the user to guide the separation process through the release of rafts from the microwell array. Start, move the raft via the magnetic wand to the collection plate, and map the collection plate Track the progress of the separation process by showing the time to completion and each deposit This makes it possible to do so.

[0100] When the system 100 is used in cytometric image analysis mode, elements similar to those in FIG. An "Imaging" tab GUI containing the following will be displayed, allowing the user to Fully automated scanning of the microwell array by navigating the system's field of view within the array In order to set the imaging parameters of the three fluorescence imaging channels used in real time Allows the user to visualize images and pixel histograms. Initiate a full scan of the ray, track the progress of the scan, and automate the entire microwell array. Allows you to pause / resume dynamic scanning in Cytometry Image Analysis mode The Array Scan tab GUI can be displayed. The GUI displays the current view in the array. The location of the field and the image acquired from that field are displayed.

[0101] FIG. 10 is a screenshot of another exemplary screen of a graphical user interface. Figure 10 shows the "cell gating" or cell gating in the cytometry image analysis mode. The Identification tab GUI shows the identification of cells on each raft in the array. To determine the number of fluorescent objects in the primary imaging channel, This control allows the user to set the parameters that are displayed on the object. Set the intensity threshold for object segmentation and enable setting changes to the segmentation. We investigated the effect of rafts on the function of single-cell rafts and further screened candidate rafts from the population. These settings allow the user to apply various filters within a single The identification of cell rafts in a single system can be optimized to achieve an efficiency of over 95%. The method 800 for single cell detection can be applied to the user interface. The direction represents pixel intensity, and the Y axis represents the number of pixels in the entire array scan at a given signal intensity. A histogram of the signal in the "cell discrimination" channel, which represents the number of cells, is displayed. Use the mouse to set the current single cell gating threshold index on the histogram. can be done.

[0102] The user interface allows the user to set a single cell gating threshold for counting on the array. The minimum (0%) and maximum (1%) single cell counts detected during the entire array scan are shown, as well as the confidence interval for the number of single cells detected. The normalized percentiles calculated as a function of both the maximum pixel intensity (100%) and the maximum pixel intensity (100%). The GUI screen also allows the user to change the value to represent all addresses of the microwell array. The total number of rafts identified as containing cells during the ray scan and the number of rafts used for subsequent isolation The number of single-cell internal rafts that meet user-specified threshold parameters is presented to the user.

[0103] The user interface now includes a "Marker Gating" option in the Cytometry Image Analysis mode. After you have completed the cell gating process, you can display the "Select Cell Gating" tab. Additional views of the interface include highlighted positions indicating the location of single cell internal rafts, and An image of the entire microwell array along with the total number of single cell-internal rafts found on the array. The user interface screen also displays two sizes specified by the user. For each of the tometry marker channels, each unit is plotted along the X-axis and Y-axis. A two-dimensional histogram can be displayed, providing a single point representing a single cell raft. may further be modified by the user in any suitable format, for example as described above with reference to FIG. A user interface that allows you to apply cytometry marker gates on the histograms The present invention can provide interface control.

[0104] The user interface also allows the user to use the mouse to move all active highlights. As well as observing images of selected single-cell rafts in the optical channel and bright field We then located the intracellular rafts of single cells within the entire image of the microwell array or on a 2D scatter plot. The interface also allows the user to click mode), the user can select a single Allows random selection of cell rafts. Plate mapping data shown in Figure 9. Using a plate mapping display similar to the 3D display, the user can visualize the selected Assigning single-cell rafts to specific locations within the collection plate identified at the beginning of the experiment This can be done.

[0105] The systems and methods described herein identify cells through their time-varying fluorescent signatures. This is applicable to many questions in cell biology where intracellular processes are observed. The effects of specific transgenes, siRNAs or small molecules on survival can be easily assessed in model cell lines. These can be easily observed and result in abnormalities such as extreme gene expression, cell proliferation or prolonged survival. Cells that exhibit the property can be isolated for expansion and more detailed study. In the field of biology, the system has been shown to be able to detect and express specific markers in primary human tumors containing diverse cell populations. It can be used to isolate and study individual cells with specific temporal characteristics.

[0106] For example, time-lapse analysis can be achieved using features inherent within the cytometry image analysis mode. For applications involving the administration of drugs or reagents, 1) single cell roughness within the array can be achieved. and 2) identifying baselines in one reporter fluorescence imaging channel. To quantify the fluorescence levels, the entire array is scanned at t=t1 before any drug or reagent is added. Addition of drug or reagent and sufficient incubation time can then be performed. A second full array scan can then be performed at t=t2. The system then: The difference in fluorescence measured between t1 and t2 is calculated and displayed, and the user can select the Allows for the selection of single cell rafts for isolation by allowing the generation of gates to.

[0107] For this workflow, where fluorescence values ​​are compared between two different time points, various permutations have been If the two time points are minutes or hours apart, the cells may remain inside the device. If the two time points are several hours or days apart, the microbes containing the cells may The cell array was removed from the instrument between imaging runs and placed in a standard cell culture incubator. It can be put in.

[0108] An additional application of two-point time course analysis is the transfection of cells with fluorescent reporter genes. The preliminary scans were performed to evaluate the growth of clonal colonies during the experiment. The secondary scan identifies the single cell rafts required for the colony selection. Total colony fluorescence, which indicates the effectiveness of transfection ) would allow for the evaluation of

[0109] To further enhance the time passing capability, the system can be configured to track the time t=t3, t4, …, t N in Any number of additional array scans are performed to quantitate the fluorescence intensity in the reporter imaging channels. A plot of the measured intensity versus time may be displayed. As previously described, single time point fluorescence intensity or two time points may be displayed. In addition to creating a single cell raft population gate based on intensity changes between time points, The software also allows users to create gates based on the shape of the fluorescence time curve. The curve shape based gating function is useful for dimensionality reduction, clustering, and principal components. analysis, Fourier transform, wavelet transform, hierarchical cluster analysis, or linear discriminant analysis Any classification technique can be used to facilitate this.

[0110] Although fluorescence intensity is an excellent measure of the abundance of an analyte within a single cell, There is also potential to use this capability to study the localization of proteins. For example, many membrane proteins Proteins are typically internalized via endocytosis and transported to other intracellular compartments such as the nucleus. They respond to a given ligand or other stimulus by translocating to a target cell or organelle. Proteins that exhibit this behavior can be analyzed not only by observing their fluorescence intensity, but also by analyzing their fluorescence signal. It may also be necessary to observe changes in the signal localization patterns. For example, If so, compare the plasma membrane-associated (-linked) signal of the protein with its nuclear-associated signal This can provide an indication of the overall signaling through a given pathway. Many of these include the organelles themselves, such as the endoplasmic reticulum, endosomes, or mitochondria. Many types of analytes can utilize this method. For these types of time course experiments The image analysis algorithm preferably employs a segmentation-based method to To identify targets and predict the nature of the subcellular compartments in which they are localized, Either their geometric shape can be input to the computer or the user can select the shape of the nuclei for a given subcellular compartment. This allows one to ascribe a specific fluorescent signal.

[0111] 11A to 11I show an actuator, an imaging device, and an image processing and analysis algorithm. An automated computer-controlled microwell release and The ELISA and collection system was used with the microwell array to detect transfected K562 cells. Here is an example of an experiment to detect and isolate a human myeloid leukemia cell line (a human myeloid leukemia cell line). A plasmid containing a gene encoding the enhanced green fluorescent protein (EGFP) reporter gene. The gene and one of two sgRNAs designed to bind to a specific sequence in the U2AF1 gene. In addition, single-stranded oligodeoxynucleotide repair templates containing the leukemia-associated S34F mutation Cells were transfected using standard techniques including transfection. 24 h after transfection, in 2 mL of pre-equilibrated culture medium at 37°C, 5% CO The transfected cells were then cultured in 1 μM calcein AM or Ce. The cells were stained with IlTracker Deep Red for 30 minutes, washed, and then restained for 10 min. Single cells were imaged according to the color protocol and seeded onto the microwell array. To identify the location of cell rafts with more than one Cell rafts containing untransfected cells were excluded from further analysis. Arrays were imaged every 12 hours until 72 hours post-transplant.

[0112] The components of the automated system are similar to those of the system 100 of FIG. The imaging device of the system was the ORCA-Fl, which was used to acquire bright-field and fluorescent images. MVX10 with ash4.0 CMOS camera (Hamamatsu, Bridgewater, NJ) Includes a MacroView upright microscope (Olympus, Center Valley, PA) It was a plan Apocalomat (pl An apochromat objective (1x with a numerical aperture of 0.25) was used during imaging. The PS3H122 electric fork has a wide range of effective magnification (0.63x to 6.3x). A gas actuator and H138A motorized XY translation stage (Prior Scientific Inc., Rockwell Automation, Inc.) The specimen and objective movement was automated using a Lambda 1 0-3 Optical Filter Changer, Emission Filter Wheel (LB10-NWE), SmartShut Excitation filter wheel with shutter (LB10-NWIQ) and stand-alone SmartShutter shutter The detector was equipped with an IQ25-SA (Sutter Instrument, Novato, CA). Five excitation bands Pass (350±50nm, 402±15nm, 490±20nm, 555±25nm, 6 45±30nm), and four emission bandpasses (455±50nm, 525±36nm, A filter set (89000-ET- Sedat Quad Chroma Technology Corp, Bellows Falls The arc lamp (VT) enabled fluorescence measurements at blue, green, red, and far-red wavelengths. Lumen 200, Prior Scientific Inc., Rockland, MA) was used for illumination. All microscope equipment was developed using custom software written in MATLAB (MathWorks, Natick, MA). The system was controlled by Micro-Manager (Open Imaging, San Francisco, CA) Core was used.

[0113] At various times, bright field and fluorescent images of the microwell array were acquired. Full image range To ensure at least 300 µm between imaged fields of view (cell rafts + cell rafts An overlap of 10 μm (interval width) was used in all experiments. For the experiment, cells were first stained with CellTracker Deep Red. Bright field and fluorescent images of B were taken at 12:00, 24 and 72 hours after transfection. were taken at intervals.

[0114] The system automatically acquires brightfield and fluorescent array images at specified times. A custom MATLAB program was developed to control the microscope and to process and analyze the images. A graphical user interface (e.g., similar to GUI 126 in Figure 1) was used. The user interface includes the fluorescence channel selection, camera exposure time, and microwell array geometry. The user interface also allows for user input of the four It is also used to prompt the user to manually position the corners and center to achieve focus. Based on the interpolated plane from the fit based on the array size, the array To predict the position and focal plane for each field of view, five identified The points were fitted to a thin-plate spline.

[0115] At all magnifications, cell rafts have high contrast boundaries under bright field illumination. Cell rafts were segmented and assigned array locations using brightfield images. A flat-field correction was performed on each bright-field image to correct for non-uniform illumination. Each bright-field image was thresholded using Otsu's method, and pixels were scored as either above or below the threshold. Based on their values, they were assigned a 1 or 0. To exclude, the binary images were further processed to fill the interior of each cell raft boundary. Objects larger than 1.5 times or smaller than 0.5 times the size of known cell rafts were identified. Using this strategy, the locations of all cell rafts were determined at each time point and across cell rafts. The background was identified before the separation. The band noise was removed from the fluorescence images. Then, each image was thresholded and the images were binarized. To convert the number of contacting cells to the number of fluorescent cells, the method of Otsu was used. A watershed algorithm was applied to the binary images to allow for the The spatial resolution required is determined by the pixel size, the array image time, and the cell raft segmentation. The method was optimized by considering the accuracy of the alignment, the accuracy of the alignment, and the success of cell discrimination.

[0116] The release probe in the automated system is made up of two Delrin components: a small stepper Linear Actuator (20DAM10D2U-K; Travel 15mm; Portes cap) from motor housings and needle mounts for West Chester, PA The needle mount has a small hole through which the needle is secured. The needle had a clear polycarbonate window. It is possible to perform bright field microscopy while the sample is placed in a specified position. The controller was an Ardui with a motor shield (Adafruit Industries, New York, NY). A custom MATLAB program connected to the Uno (SparkFun Electronics, Boulder, CO) The target cell raft list was generated by image analysis software. A customized protocol written in MATLAB to release individual cell rafts when fed An automated machine system has been developed having software that is A stepper linear actuator punctures the microwell array and separates the individual wells. Z-position of the release probe used to move cell rafts within individual microwells The release probe and motor were connected to a microscope objective aligned with the optical path of the microscope objective. The user interface was attached under the microscope stage and was used to control the release of cell rafts. Allows the user to select the Z travel distance during operation. The cell rafts formed depended on the number of times the release needle penetrated the microwell array at the target position. The results showed that the ions were successfully removed with an efficiency of 94–100%.

[0117] A motorized magnetic wand in an automated system captures, transports, and transfers superparamagnetic cell rafts. It was designed to be placed inside the collection vesicle. The collection wand is fitted with a cylindrical NdFeB magnet (diameter 3. 175 mm, length 25.4 mm) into a hollow polycarbonate cylinder (outer diameter 4.76 mm The magnet was fabricated by placing it inside a magnet (magnet with a diameter of 3.18 mm and a length of 63.5 mm). The cylinder is blocked at both ends so that the is free to move along the central axis of the cylinder. The retrieval wand was attached to the microscope objective using a Delrin part and was aligned perpendicular to the objective lens. The directional movement was achieved by a linear actuator (L12-30-50-06-R; Figelli Technologies Inc., The linear actuator was controlled at 30 mm travel (Victoria, BC, Canada). Controlled by a custom MATLAB program interfaced with an Arduino Uno equipped with a motor shield To capture the cell rafts, the magnetic collection wand was placed 2 m away from the released cell rafts. The cells were placed in the medium above the microwell array, within 100 μm of each other. Once removed from the array, The cell rafts are formed not only by the surface tension of the fluid droplet on the tip of the wand, but also by the circular The cell rafts were held on the tip of the wand by a cylindrical magnet. Placed on a plate.

[0118] Before retrieving the cell rafts from the array, the position of the release needle within the field of view was recorded. The XYZ position of the collection wand relative to the ray is also relative to the position of the wells in the 96-well plate. The relative positions of the rafts were also calibrated. The selected cell rafts were identified by needle piercing the PDMS substrate using a microfluidic device. The magnetic collection wand was immersed in the array culture medium and released. The magnetic cell rafts were attracted. The collection wand was then placed on a 96-well plate containing culture medium. The plate was moved to a well near the plate and immersed in it. A NdFeB block magnet (101.6 × 76.2 × 6.35 mm) was used to organize the cell rafts. The block magnet also attracted the cylindrical magnet in the collection wand to the block magnet. It was positioned to be repelled.

[0119] Customized MATLAB to integrate the operation of the various components of the system A program and user interface was used to image the cell raft release probe. Coordinated migration, raft release, collection of cell rafts, and cell culture of the collected cell rafts on a 96-well plate The user interface allows the user to move the microscope stage to place the sample within the range of the rate. Using this, the user begins cell raft release until the needle just breaks through the PDMS. The user identifies the position of the needle and the software adjusts the travel distance of the XY stage. The GUI then prompts the user to adjust the XY stage and the cell The raft collection system was manipulated to align the wand tip at the four corners and 96 wells of the cell raft array. The cells were then placed in the four corner wells of the plate. This information is then used to interpolate the position.

[0120] Identify the location of cell rafts with single cells and measure the duration and time of EGFP fluorescence in each cell. An automated system was used to track the time and intensity of EGFP-positive cells over a 72-hour period. At any given time point, cell rafts containing EGFP-expressing cells were examined every 12 h. The image analysis software identifies when the imaging time course is complete. A total of 220 cell rafts contained at least one fluorescent cell during the imaging time, which corresponds to 1 This corresponds to a transfection efficiency of 0.9%. Cell rafts containing fluorescent cells were 100% Sensitivity: The automated cell raft system liberates cell rafts with >98% efficiency, and Cell rafts could be collected with 0% efficiency. Release and collection into 96-well plates For this purpose, single cell rafts were targeted. Cells expanded into colonies were genetically analyzed. The presence of successful gene edits was determined. Two genes containing the S34F mutation in U2AF1 were identified. K562 colonies were generated and cells were sorted based on the temporal evolution of fluorescent protein expression (s ort) and provide selected cells and colonies in which gene editing was successful. Ta.

[0121] Time evolution of EGFP expression in transfected cells on the microwell array (te mporal evolution, and EG predicts whether cells will proliferate after detachment of cell rafts An automated system was also used to analyze whether there were any clear trends in FP expression. We classified individual cells on cell rafts into groups with low fluorescence throughout the entire period, high fluorescence throughout the entire period, low expression followed by The imaging system was used to categorize the expression as high expression followed by low expression, and high expression followed by low expression. The mean fluorescence per cell obtained by the system over 72 hours was used. The expanded cells were distributed stochastically across the four groups. There was no discernible difference in fluorescence expression at any single time point between the proliferation groups and the microwells. There was no correlation between cell proliferation on the arrays and proliferation after cell raft isolation. EGFP expression Time evolution analysis of transfected K562 cells was also performed They demonstrated that it can take up to 72 hours for the cells to subsequently express EGFP. The system allows users to collect expression data over time in single-cell rafts in microwell arrays. This allows the data to be correlated with downstream molecular analysis, in this case separation chromatography. The determination of successful gene editing in a single cell was assayed on a microwell array. Derived from a single cell.

[0122] FIG. 11A shows the identification and isolation of cell rafts containing transfected K562 cells. Figure 11B shows a schematic of the image processing and analysis used to measure the microwell alignment. Figure 11C shows the raw brightfield image of the ray after flat-field correction has been applied. Figure 11D shows a bright field image of the same microwell array. Figure 11E shows the thresholding of the corrected image, generating a binary image marking cellular rafts. applied to remove any cell rafts that fill the interior and touch the border of the image. Figure 11F shows morphological filtering. Figure 11G shows a fluorescent image of two contacting cells after the experiment. Figure 11H shows a top-hat filter applied to the fluorescence image to detect the adhesion of two cells. Figure 1 shows thresholding of the top-hat filtered image, resulting in the concatenated binary image. 1I shows that the Watershed algorithm was applied to separate contacting cells. The results are shown below.

[0123] In one embodiment of the present invention, to investigate the simultaneous measurement of individual T cell recovery and T cell function, It includes an actuator, an imaging device, and a processor that contains image processing and analysis algorithms. An automated computer-controlled microwell release and collection system is provided for the microwell array. was used in conjunction with

[0124] In functional assays on arrays, individual T cells react with target cell populations. The ability of the T cells to kill the T cells was measured, and the time dependence of T cell-mediated killing or cytotoxicity "on the array" was measured. To measure the solubility of the cellulose, an automated system (similar to system 100 in FIG. 1) was used. The system further allows for the observation of antigen presentation under different conditions and on different cell types. Specific cells were then sorted into 96 sections to determine the clonal population that could be used as a probe for was used to classify well plates as viably.

[0125] Human T cell cultures were generated against influenza M1p antigen using standard methods. D8 + To obtain the cells, isolated peripheral blood mononuclear cells were used. Cytotoxic T lymphocytes ( CTL cultures were performed in 10% human AB serum supplemented with IL-21 (complete medium). ia), CM) in CTS AIM V medium. + T cells were co-immunopreserved with M1p-pulsed DCs. After 3 days of culture, the cells were treated with IL-7 every 2 days. and IL-15. Eleven days after the initiation of culture in CM, CTLs were restimulated with M1p-pulsed DCs. IL-2 was added 19 days after the start of culture. CTLs were restimulated on days 21 and 34 after the start of culture. The cells were then frozen and stored in aliquots on the 41st day after the start of culture. TLs were thawed and restimulated with M1p-pulsed DCs. After 3 days, CD8 + T cell isolation kit CTLs were isolated using the IL-7, IL-15 and IL-2 supplemented CTLs. After 2-3 days, CD8 + T cells were immobilized on a microwell array M1p / HLA-A * 02:01 Tetramer enumeration is 48.4% CD8 + The bulk cultures showed that T cells were specific for M1p. Antigen-induced hypersensitivity to autologous DCs pulsed with M1p compared with autologous DCs pulsed with PR1. It showed specific cytotoxicity.

[0126] Dendritic cells (DCs) were cultured using the CD34 MicroBead Kit UltraPure (Milt The frozen tissue samples were obtained from the Hematopoietic Progenitor Cell Laboratory at UNC Hospital using Enyi Biotec. CD34 isolated from stored leukapheresis products + Distinguished from cells The method for obtaining cryopreserved immature DCs for future use is described in Supplementary Methods. As described above, 10% was supplemented with GM-CSF, Flt3-ligand, SCF, and IL-4. CD34 in CTS AIM V medium containing human AB serum (complete medium, CM) + Cell 1 Immature DCs were cultured for 2 days. Immature DCs were cultured for 2 days with GM-CSF, IL-4, and TNF-α. The cells were then cultured for 2 days with GM-CSF, IL-4, TNF-α, IFN-α, and IL-6. The mature DCs were differentiated by injecting the peptides (M1p or PR1) into the DCs. The cells were then co-cultured for at least 18 h.

[0127] The microwell array was coated with 0.1% by weight bovine gelatin in PBS and incubated for 3 h. The cells were incubated at 7°C for at least 2 hours, after which the gelatin solution was aspirated and the array was removed before fixing the cells. The sections were washed three times with PBS.

[0128] The microwell arrays used in the analysis were used to culture cells and then visualize them over time using a microscope. The raster consisted of a regular pattern (70 × 70) of cell rafts that were dynamically assayed. Cell rafts fabricated from magnetic polystyrene were designed to protect the cells from the elements during assay setup and execution. The microwell array has a concave surface of 120 μm depth that enhances cell retention on the surface. were pulsed with either M1p or PR1 peptide and labeled with Hoechst dye. Autologous DCs (the “target” cells) were seeded. The target cells were stimulated with human AB serum, penicillin, streptavidin, and serovar IgE. Leptomycin and HEPES were added to the cell rafts at a ratio of 30 cells per array. 147,000 cells) in phenol red-free RPMI 1640 supplemented with The CD8 + T cells with CellTracker Deep Red Cell:cell raft ratio to maximize the number of cell rafts containing labeled single T cells The cells were placed on the array in a 1:1 ratio. The number of cells in each well across the array was proportional to the Poisson distribution. The stochastic deposition of the ions onto the array follows the fabric, which covers approximately one-third of the wells (36.8%). or 1803 wells) are single CD8 + predicted to have T cells, 26.4% (12 94) were predicted to contain more than one T cell, and 36.8% (1803) contained no T cells. The medium covering the array is a membrane-impermeable DNA-binding dye. The antibody contained Sytox Green, which is a stimulant for the treatment of inflammatory bowel disorders.

[0129] The assay involves single T cells and measures increasing green fluorescence over the course of 6 hours. The study was designed to identify cell rafts that exhibit high cytotoxicity against target cells expressed by Both target cells and T cells were applied to the medium on the microwell array. Thus, individual cell rafts can harbor different numbers and ratios of target and T cells. The acquisition, processing, and analysis are performed in a manner similar to that described above with reference to system 100 of FIG. Fluorescent images of the arrays in the incubator housing the microscope were taken over a 6-hour period. The development of green fluorescence over time for each cell raft was measured every 30 min. was measured.

[0130] The fluorescent images were processed in a manner similar to that described above with reference to Figures 4A to 4E. Fluorophores (Hoechst, Sytox Green, and CellTracker Deep Red) were expressed. Fluorescence images were analyzed to determine the intensity, area, location and number of cells. To generate a binary mask of cells on the cell array, a top-hat filter was used. Filtering and Otsu's thresholding were applied to each image. Within the mask, the intensity, position and pixel count were recorded for each cell raft. To count the number of individual cells on rafts, we used the far-red channel (CellTracker Deep Red). The watershed algorithm was applied to each image in (corresponding to a stained cell). .

[0131] In parallel with the image acquisition, a MATLAB program processed and analyzed the acquired images. A MATLAB GUI is used to select the desired combination of brightfield and fluorescence channels for The processing time for each set of images was faster than the microscope stage movement and image acquisition. , no additional time was added to the total scan time. Identifying the location of individual cell rafts Because of the elastomeric nature of PDMS, the cell laminae on the array were not visible. Image analysis was required to pinpoint the exact location of the rafts. The method of annotation includes background estimation, flat-field correction, thresholding and shape correction. The majority of cell rafts (99.8 ± 0.8%) were Positive but not correctly identified (n = 100 images, 100–121 cells per image raft).

[0132] The analysis performed on the microwell array showed that 0 or 1 CD8 + Either T cells The time course of green fluorescence in cell rafts containing T cells was compared with that in cell rafts not containing T cells. The control was considered as a control and was reflected in the rate of spontaneous target cell death over the study period. A total magnification of 4x was used to scan the cell rafts, with a resolution of 1.62 μm / pixel. This magnification was chosen to minimize image acquisition time. This allows for easy identification of single cells while maintaining a wide field of view to facilitate imaging. Each 70 × 70 microwell array captures one channel at this magnification with 5% image overlap. 49 images were required per channel. Brightfield (100 ms camera exposure) and 3 fluorescent Image acquisition of a single array using a single channel (each 200 ms camera exposure) takes The time required for the image to be captured was 216 ± 4 seconds (n = 10). An algorithm was used that took 99±2 seconds to complete. All focal points were automatically acquired. The planes were imaged at 4x magnification (±21.8 μm) compared to manually selected focal planes (n=50). The imaging was accurate within the depth of field of the microscope objective. Scanning time (autofocus and image acquisition) was 315 ± 5 s.

[0133] The positions of individual cell rafts were determined using image processing and analysis software. CD8 + Cell rafts containing cells were identified using CellTracker Deep Red fluorescence. These cell rafts were classified based on the increase in Sytox Green fluorescence intensity. Cell rafts were automatically identified to ensure that Deep Red-positive cells remained on cell rafts. The samples were rescreened after gelatin encapsulation.

[0134] The automated system and process detects fluorescently labeled antigen-presenting cells, as measured by fluorescence microscopy. A population of target cells and one CD8 + A fixed number of cell rafts were designed to contain T cells. This allows the measurement of the rate of T cell-mediated killing of single cells. , highly cytotoxic CD8 in just 2 hours + It is now possible to distinguish T cells. Cell killing was measured by the Sytox Green pixel intensity within the Hoechst fluorescent area on each cell raft. The Sytox Green fluorescence was measured by summing the total fluorescence intensity of each cell raft. Using the Hoechst positive area, debris and disintegrating cells and These fragments correspond to the specific number of effector cells present. To generate a temporal trace of cytotoxicity, effector cells were assayed for each cell raft at each time point. The number of CD8 cells and cytotoxicity information were recorded. + Target cell death in T cells The rates varied widely, however, and individual T cells responded well over the time course of the experiment. maintain their cytotoxicity rate, which is a high cytotoxicity of CD8 + Rapid T cell This allows for easy identification.

[0135] Once the cytotoxicity assay was completed, the arrays were covered with a thin layer of gelatin as previously described. The incubator surrounding the microscope was cooled to 24°C just prior to gelatin covering of the array. The culture medium on the array was replaced with 5% by weight bovine gelatin in PBS, and the array was centrifuged. The arrays were then incubated at 37°C for 10 minutes, the excess gelatin was aspirated, and the cells were then The rafts were incubated at 4°C for 5 min to solidify the gelatin within the rafts. The gel was then layered on the array. The cell lamination with one or more gel-encapsulated cells was then performed. The probe is easily captured by a magnetic wand placed over it and immersed in the medium above the array. The wand with the captured cell raft was then used to move the cell raft to the bottom of the well. Within the wells of a 96-well plate, facilitated by a magnet underneath the plate to pull was placed in.

[0136] CD8 + Cell rafts, which contain T cells and show a high percentage of target cell death, were identified as needle release-associated The rafts were then individually released from the array using a computer-controlled The magnetic field is collected by a magnetic wand attached to a three-axis motor and is described above with reference to FIG. The needle solution attached to the microscope was placed in a 96-microwell plate in the same manner as in the previous experiment. Using a release device, highly activated CD8 + Cell rafts with T cells contain feeder cells The three sorted T cells were then clonally expanded into wells of a 96-well plate. All three clones were M1p / HLA * 02:01 High binding activity to tetramers (avi dity) T cell receptor expressed.

[0137] This method involves the use of clones known to be cytotoxic to a selected target cell. The data can be used to determine the sequence of the T cell receptor (TCR) in It can be used to develop immunotherapies. TCR consists of a TCR α chain paired with a TCR β chain. They exist as heterodimeric protein complexes that are expressed on the surface of T cells. The paired chain is, in this case, M1p / HLA * 02:01 Complex, target peptide Binds to target peptides / MHC with high affinity. Confers specificity to the interaction with target peptides / MHC. The portions of the TCRα and TCRβ chains that bind to each other are known as CDR3 regions. If the CDR3 region of the Rβ chain and the adjacent V and J segments are known, the full length (fu A ll-length transgenic TCR construct was generated and the target peptide / M Transfecting primary human T cells to alter their specificity for HCs This approach targets TCRα and MHC complexes specific for cancer. T cells from cancer patients were transduced with a transgenic TCR containing the TCR β chain sequence. It may have broad applications in cancer immunotherapy, where it can be highly The ability to determine the activity of small numbers of effector T cells on short time scales may also be valuable in the development of T cell therapy. As they are being developed for clinical use, they may also be used for immune monitoring. Such assays are useful to assess the off-target cytotoxicity observed in early T cell therapy studies. To reduce the risk of seroconversion, T cell cultures are screened for antigen specificity against multiple antigens. In addition, modification of the geometry of the array can be used to It can be easily modified to identify cytotoxic T cells, which are rare tumor antigen-specific CD4 T cells. 8 + This would allow the identification and subsequent cloning of T cells. The assay is used to investigate the resistance of target cells to T cell-mediated cytotoxicity, or to Screening for drug candidates that modulate the activity of multiple interacting cell populations simultaneously can be used to examine the effect of

[0138] Processors, actuators, cameras, motors, magnets, and other components, and implementation The possible codes may form various means for carrying out the embodiments of the present invention. In some embodiments, the processor may be a DSP, a microcontroller, or a microprocessor. What general purpose processor is used and what non-transitory firmware, software, or microprocessor is used? The code may be stored in a tangible storage device associated with the system. computer, mobile computer, or embedded controller-based system Any such embedded computer can be connected to other hardware. Such functionality may be referred to herein as a "processor" or "microprocessor." Storage may be memory integrated with a processor or memory that is used by a processor to perform control functions. The memory chip may be addressed by a controller or processor. Such firmware, software or microcode may be executed by a processor. Such firmware can be used to execute a program that, when executed, causes a processor to perform that control function. The hardware or software may also be connected to or within the cell separation system. Optical disks or conventional removable or fixed magnetic media such as disk drives It can be stored in or on any device.

[0139] Any software necessary to support the execution of instructions for any embodiment of the present invention The software and any data and information may be removed for development purposes or for maintenance and update purposes. Note that the data may be stored on a functional storage medium. Such storage may be either directly or can be accessed over a network, including the Internet.

[0140] Although specific embodiments have been shown and described herein, those skilled in the art will appreciate that various modifications may be made to achieve the same objectives. It is understood that any of the above contemplated configurations may be substituted for the specific embodiment shown, and that the present invention is not limited to the specific embodiment shown. It will be understood that the present invention has other applications in various environments. The appended claims are in no way intended to cover the scope of the invention. The present disclosure is not intended to limit the scope of the present invention to the specific embodiments described herein.

Claims

1. 1. A system for releasing cell rafts from a microwell array, comprising: Imaging configured to obtain an image of a microwell array containing a plurality of cell rafts Device; an actuator configured to release cell rafts from the microwell array. and 1. A computer system comprising at least one processor and a memory, comprising: The computer system comprises: acquiring one or more images of the microwell array using the imaging device; Analyzing the one or more images of the microwell array identifying a selected cellular raft from among the plurality of cellular rafts; and and releasing the selected cell rafts from the microwell array. controlling the controller; The system is programmed to automatically image and release cell rafts using 。

2. 2. The system of claim 1, configured to move the imaging device, the microwell array, or both. one or more motors formed in a Acquiring the one or more images of the microwell array comprises: The microwell array is divided into a plurality of fields, and the fields are each divided into a plurality of cell rafts. and for each field of view, directing the imaging device to the field of view of the microwell array; and controlling the one or more motors to capture respective images in the field of view. Controlling the imaging device; performing an array scan by

3. 3. The system of claim 2, a microscope objective for said imaging device; Acquiring one or more images of the microwell array includes, for each field of view: using one or more focus positions from at least one adjacent field of view to image the microscope pair; a system for autofocusing an objective lens.

4. 4. The system of claim 3, Autofocusing the microscope objective includes performing a plurality of focusing operations at a plurality of sample focus positions. and sampling the sample image of the current focus score from the sample image. and interpolating the current focus position.

5. A system according to any one of claims 1 to 4, Identifying the selected cellular rafts comprises, for each cellular raft, Count the number of cells depicted in the sub-images and identify single isolated cells. and identifying at least one single-cell raft.

6. 6. The system of claim 5, Counting the number of cells depicted in the sub-images is performed prior to creating a binary image. applying an intensity threshold to the sub-images to identify unique objects in the binary image. and counting the number of unique objects identified. Hmm.

7. 7. A system according to claim 5 or 6, comprising: Identifying the single-cell rafts may include determining whether the single-cell rafts are present in the single, isolated cell. determining a confidence score indicative of a confidence in the decision to accommodate the cell. Tem.

8. A system according to any one of claims 1 to 7, Identifying the selected cellular rafts comprises, for each cellular raft, detecting a marker depicted on the sub-image; and determining whether the cell lane is located based on the detection of the marker. and gating the buffer.

9. A system according to any one of claims 1 to 8, Identifying the selected cell rafts includes collecting the selected cell rafts. The system includes assigning the address to a mapped location.

10. 10. The system of claim 9, A mechanical cell raft collector is provided. The computer system detects the mechanical defects after release from the microwell array. controlling a cell raft collector to collect the selected cell rafts; and The raft collector is controlled to deposit the selected The system is configured to deposit the deposited cell rafts.

11. 1. A computer-implemented method for collecting cells, comprising: The information on the microwell arrays containing multiple rafts was analyzed using an automated computer storing the data in a system memory; performing image analysis using a processor of the automated computer system; and and using the information to identify cellular rafts from among the plurality of rafts; and The processor and the actuator are used to extract the cell laminae from the microwell array. Releasing the soft.

12. Storing the information regarding the microwell array further comprises: Identifying and storing the size of the microwell array; Identifying and storing an optimal focus position and an optimal exposure for said microwell array. ； A processor is used to section the microwell array and generate a set of microwells.

12. The method of claim 11, further comprising storing the movement required to align the cross well with the field of view. A computer-implemented method of.

13. determining an address for each of a plurality of microwells in the microwell array; and calculating and storing an offset of the actuator.

13. The computer-implemented method of claim 12.

14. 1. A system for monitoring intracellular processes, the system comprising: A camera configured to capture images of a microwell array containing a plurality of cell rafts. An imaging device; a computer system comprising at least one processor and a memory; The computer system: first, using the imaging device, determining that the cellular raft comprises one or more cells undergoing an intracellular process; The at least one cellular raft is undergoing the intracellular process. In response to determining that the at least one microwell on the microwell array contains more than one cell, memorizing the location of another cell raft; and At least one cell rough using said location one or more times after said first time. Monitor intracellular processes by identifying the location of the target and by imaging the target with a small acquiring one or more images of at least one cell raft; The system is programmed to automatically image and release cell rafts using 。

15. Monitoring the intracellular process includes storing the one or more images; and storing each of the one or more images in a time and date format corresponding to each of the images. The method of claim 14, further comprising: associating said at least one cellular raft with said at least one cellular raft. Tem.

Images