Systems and methods for sorting laser particles or cells

Abstract

The flow sorting system and method include a sample loader configured to receive a sample containing one or more laser microparticles, each laser microparticle being configured to generate laser radiation having one or more different spectral peaks when excited. The system further includes a spectrometer that receives the laser radiation from the one or more laser microparticles and generates spectral data, and a processor configured to receive the spectral data and generate a sorting signal. The system also includes a switch configured to receive the sorting signal and route the one or more microparticles to a particular one of a plurality of collection channels based on the sorting signal.

Description

Technical Field

[0001] Cross - Reference to Related Applications This application claims priority based on U.S. Provisional Application No. 63 / 339,491, filed on May 8, 2022, and the entire disclosure thereof is incorporated herein by reference for all purposes.

Background Art

[0002] The present disclosure relates to systems and methods for sorting used laser particles or cells. More specifically, the present disclosure provides systems and methods for identifying microparticles that provide laser radiation, which can be used, for example, to sort microparticles or cells physically associated with the microparticles based on the laser radiation characteristics of the microparticles.

[0003] Flow sorters are tools widely used in various applications including life sciences, particularly cell analysis. Sorters are typically integrated with flow cytometry to analyze the biochemical and physical properties of cells or microparticles. Usually, the properties are measured using fluorescence signals from individual cells. Various active and passive sorting mechanisms are known, such as electrostatic droplet - based sorting, size - based separation, inertial separation, etc. This sorting technology is commonly referred to as Fluorescence - Activated Cell Sorting (FACS).

[0004] FACS is the most common type, which transfers and collects cells into various vials based on fluorescence signals related to immunostaining biomarkers, survival rates, reporter gene expression, etc. Light scattering-based sorting is also used for label-free purification of cell subpopulations such as lymphocytes and pancreatic islet cells. Magnetic-activated cell sorters (MACS) are also increasingly used for the purification of genetically modified cells for cell therapy. Fluorescence image-based cell sorting has been recently developed to selectively isolate single cells with unique spatial and morphological characteristics. In all of these technologies, the sorting signal or routing decision is related to the cell phenotype measured in situ.

[0005] Laser particles (LPs) are micron- or submicron-sized particles that can generate induced laser emission. By optical excitation or pumping, each LP emits radiation with a spectral linewidth of sub-nanometers, providing an ultra-pure and distinguishable laser color. Recently demonstrated semiconductor microdisk lasers are representatives of LPs. When the wavelength interval is 1 nm, approximately 400 colors in the range of 1200 - 1600 nm are realized, and in principle, it can be extended to millions of barcodes using combinations of LPs. These optical barcodes are suitable for tagging individual cells on a large scale (>10,000 cells) for single-cell analysis.

[0006] Imaging of such LPs and LP-tagged cells is achieved by using a microscope that integrates a high-resolution optical spectrometer and a pump laser. This enables the tracking of LP-tagged cells. A flow cytometer that integrates a spectrometer and a pump laser has been demonstrated for reading LP barcodes. However, a flow sorter that can sort LPs and LP-tagged cells has not been available. Therefore, a method for effectively and efficiently utilizing each of these technologies and systems has not yet been found by researchers. SUMMARY OF THE INVENTION

[0007] The present disclosure provides a system and method that overcome the aforementioned drawbacks by sorting laser particles based on their spectral emission characteristics. In one non-limiting configuration, a flow sorter is provided that is configured to route laser particles or cell entities tagged with laser particles to a plurality of collection channels, and the routing or sorting decision is made based on the spectral emission characteristics of the laser particles compared to reference data that functions as a sorting criterion. The reference data is used to collect the microparticles in one of the collection channels and can share substantially similar barcode characteristics with respect to the laser peak wavelength, but these characteristics are substantially different from the characteristics of the microparticles collected in different output collection channels.

[0008] According to one aspect of the present disclosure, a flow sorter system is provided that includes a sample loader configured to receive a sample containing one or more laser microparticles. Each laser microparticle is configured to generate laser radiation having one or more different spectral peaks when excited. The system further includes a spectrometer configured to receive the laser radiation from the one or more laser microparticles and generate spectral data, and a processor configured to receive the spectral data and generate a sorting signal. The system also includes a switch configured to receive the sorting signal and route the one or more microparticles to a particular one of a plurality of collection channels based on the sorting signal.

[0009] According to another aspect of the present disclosure, a method for sorting laser microparticles is provided. The method includes loading a sample containing one or more laser microparticles into a microfluidic system, exciting the one or more laser microparticles to generate laser radiation for each of the one or more laser microparticles, and analyzing the spectrum of the laser radiation to determine the characteristics of the laser radiation for each of the one or more laser microparticles. The method also includes sorting the one or more laser microparticles based on the characteristics as each of the one or more laser microparticles travels through the microfluidic system, and sending each of the one or more laser microparticles having predetermined spectral characteristics to a common outlet of the microfluidic system.

[0010] These aspects are not limiting. Other aspects and features of the systems and methods described herein will be described below.

[0011] The foregoing features of the embodiments will be more readily understood by reference to the following detailed description when taken in conjunction with the accompanying drawings.

Brief Description of the Drawings

[0012]

Figure 1

Figure 2

Figures 3A-D

Figures 4A-B

Figures 5A-E

Figures 6A-B

Figures 7A-D

Figures 8A-D

Figures 9A-B

Figures 10A-D

Figures 11A-D

Figure 12

DETAILED DESCRIPTION OF THE INVENTION

[0013] Before explaining the present invention in more detail, it should be understood that the present invention is not limited to the specific embodiments described. It should also be understood that the terms used herein are for the purpose of describing particular embodiments only and are not limiting. The scope of the present invention is limited only by the claims. The singular forms “a,” “an,” and “the” used herein include plural embodiments unless the context clearly dictates otherwise.

[0014] It will be apparent to those skilled in the art that many additional changes are possible without departing from the concept of the present invention other than those already described. In interpreting this disclosure, all terms should be interpreted in the broadest possible manner consistent with the context. Variations of the term “comprising” should be interpreted as non-exclusively referring to elements, components, or steps, such that the referenced elements, components, or steps can be combined with other elements, components, or steps not explicitly referenced. Embodiments expressed as “comprising” a particular element are also contemplated as “consisting essentially of” and “consisting of” those elements.

[0015] Flow cytometry is a technique that enables the rapid and continuous analysis of the chemical and physical properties of cells or particles. Since the introduction of the earliest flow cytometer equipment in the 1950s, there have been significant advancements in commercializing this technology, such as increasing the number and speed of parameters that the equipment can detect. State-of-the-art modern flow cytometers can measure nearly 30 parameters for tens of thousands of cells per second. Usually, most of these parameters are fluorescence signals associated with individual cells. Cells may exhibit specific fluorescence colors either by their inherent gene protein expression (such as GFP-labeled cells) or by the exogenous introduction of fluorescent markers.

[0016] Furthermore, by combining flow cytometry with an active sorting mechanism, it is possible to select and separate the particles or cells of the required sub-population. In the case of cells, this technique is called Fluorescence-Activated Cell Sorting (FACS) and is routinely used to separate the required cell sub-populations. Multiple sorting mechanisms are possible. One of the most common mechanisms is to electrostatically deflect cells encapsulated in droplets based on their charge and direct them to specific outflow channels. Other active sorting mechanisms using acoustic waves, mechanical valves, and magnetic forces have also been demonstrated.

[0017] To simultaneously identify multiple fluorescent colors, many high-end state-of-the-art devices need to use up to five different light sources and multiple excitation and detection paths with up to 40 - 60 photomultiplier tubes or avalanche photodiodes. Despite this complexity, the main drawback of fluorescence-based flow cytometry is that the number of colors that can be detected is limited due to spectral overlap resulting from crosstalk between different phosphors. Since the spectral bandwidth of the emission of each phosphor is wide, light leaks into multiple PMT detectors. Despite the development of computational techniques to minimize the effects of this crosstalk, the number of colors that can be measured simultaneously is limited. To analyze more parameters simultaneously, alternative techniques such as cytometry by time-of-flight (CyTOF) that are independent of color have also been developed. Instead, heavy elements are used as tags and then identified by mass cytometry. Although a large number of tags (>40) can be used simultaneously, the detection of these tags is inherently destructive and prevents the re-culturing of the measured cells.

[0018] The present disclosure provides a system and method for a flow sorter for separating cells or particles having specific characteristics. Conventional sorters make sorting decisions using fluorescence signals from a sample, whereas the disclosed sorter uses the spectral characteristics of laser radiation from "laser particles". Using the systems and methods provided herein, sorting decisions can be made based on laser signals that are different from fluorescence. Thus, using the systems and methods provided herein, laser particles can be used to represent a new type of barcode suitable for identifying and tracking individual cells for single-cell analysis.

[0019] The biological complexity of organisms arises from the diversity and interactions of individual cells. Light imaging techniques at single-cell resolution have played a very important role in deepening the understanding of cell identity and function. However, current imaging techniques, although widely used to distinguish several different cell populations, rely on fluorescent molecules with broad spectral emission that results in significant spectral cross-talk and thus cannot be extended to large-scale single cells. This disclosure presents an intracellular optical probe called a laser particle (LP) that can have a spectral linewidth of sub-nanometers. This narrow-band emission enables the generation of hundreds of unique colors suitable for cell multiplexing. An expandable method has been developed for generating billions of micron-sized LPs from a single semiconductor wafer using a top-down manufacturing approach. Furthermore, the particle design is refined by perturbing the optical mode using nanoscale scatterers to optimize the emission signal. By physically combining multiple LPs, the number of unique optical barcodes can be expanded from hundreds to tens of thousands. Using these LP barcodes, thousands of mammalian cells are tagged and their barcode emissions are read using a modified microscope and a custom-developed flow cytometer. This disclosure provides a platform for identifying single cells in various single-cell measurements and enables the integration of acquired data of the same cells based on optical barcodes.

[0020] Laser particles include classes of cell tags that can be used to enable further advanced multiplexing. As will be described later, "laser particles" (LP) can refer to microparticles that can emit coherent light when examined by appropriate excitation. The output spectrum can be individual narrowband laser lines and is usually related to its specific shape and composition. Thus, as will be described later, this functions as an optical barcode of the cell entity associated with the laser particle. As some non-limiting examples, LPs include oligonucleotides, drugs, or other molecules physically associated with the microparticle. "Microparticle" can refer, for example, to three-dimensional particles usually less than 100 μm in size. A particle of 10 nm size is still a "microparticle" in this context since it is less than 100 μm in size. "Cell entity" includes cells, or parts of cells such as nuclei, vesicles, or organelles, or organized tissues of cells such as tissues and multicellular spheroids. The cell entity may be living or chemically fixed. "Tagging" a cell entity involves physically associating one or more barcode particles with the cell entity. In the case of cells, tagging can be achieved by attaching barcode particles to the cell membrane or inserting the particles into the cytoplasm. The "physical association" between the oligonucleotide and the laser particle and between the cell structure and the cell entity can be established, for example, using structural factors. Structural factors include, for example, any of direct chemical bonds, linkers, encapsulation, and various physical constraints. Two physically associated items need not necessarily be so, but are, for example, in proximity to each other within a range of, for example, within 100 nm or, in some cases, 10 nm or less.

[0021] Upon optical excitation, known as pumping, each LP emits radiation with a sub-nanometer spectral linewidth. In 1-x Ga x As y P 1-yMultiple types of LPs have been demonstrated, including semiconductor-based microdisk LPs composed of alloys of group III and group V materials such as. In non-limiting examples, these disk-shaped particles can have a diameter of about 2 μm and a thickness of 0.2 μm. The diameter of each LP determines the emission wavelength of its laser peak. LPs of different diameters can each emit unique narrow-band emissions within a spectral gain bandwidth of 100 nm, the range of which is exactly In 1-x Ga x As y P 1-y dependent on the material composition. Therefore, multiple spectrally offset gain bandwidths can be used to further increase the number of colors. Each bandwidth is obtained from a specific combination of x and y corresponding to a specific In 1-x Ga x As y P 1-y composition. By reading the signals of the LPs in a fluid system, high-speed analysis can be performed and the analysis can be extended by manipulating the particles and related cells.

[0022] As a non-limiting example, LP-based flow cytometry capable of simultaneously detecting more than 500 laser colors will be described. A single excitation source and a single optical detection path for determining the emission color using a high-resolution custom-built spectrometer are used.

[0023] As a non-limiting example, Laser Particle Activated Cell Sorting (LACS) at speeds exceeding 1 kHz will be described. By sorting based on LP laser emission, cells containing LPs of a specific wavelength are separated, and ultimately a batch of cells containing LPs is created. In this batch, most of these tags emit laser light within a narrow ±2.5 nm spectral window, providing a level of spectral purity exceeding that achievable using fluorescence. Also described are fluidic operations of LPs within microchannels that either further enhance the reading of LPs or enable the performance of operations that manipulate LPs at high speeds.

[0024] In a non-limiting example, another embodiment is configured with a similar system, but is configured to sort laser particles without tagging cells or cell entities.

[0025] FIG. 1 shows a non-limiting exemplary schematic diagram of a dielectrophoresis-based bidirectional microdroplet sorting device 100 according to aspects of the present disclosure. The device 100 includes a sample loader 101. In one configuration, the sample loader 101 includes a droplet oil inlet 102, a spacer oil inlet 104, and a cell inlet 106, as well as a (-) outlet 108 and a (+) outlet 110. The sample loader 101 can also be configured to deliver cells suspended in a medium flowing into the cell inlet 106. Further, a recirculator 111 is configured to supply one or more laser microparticles routed to the sample loader 101.

[0026] In a non-limiting example, a fluorinated oil immiscible with the aqueous cell medium is flowed into the droplet oil inlet 102 and the spacer oil inlet 104. The droplet oil inlet 102 constricts the aqueous flow to produce aqueous droplets with a diameter of 50 μm. The spacer oil inlet 104 spaces the droplets at approximately 500 μm intervals so that each droplet is sorted independently. The central sorting zone 112 has a channel 114 with a width and depth of 55 μm and 25 μm, respectively, and an average flow velocity of 0.5 m / s. The droplets can enter the flow channel 114 at a rate of about 1.5 kHz or more, such as 2 kHz, or tens of kHz. The device 100 also includes a pair of microelectrodes 116 and 118 offset from the proximal sidewalls of the flow channel 114.

[0027] FIG. 2 shows a non-limiting example of a sorting system 200 that uses the apparatus 100 of FIG. 1. The system 200 includes an excitation source 202, non-limiting examples of which include a pump laser. The system 200 also includes a spectrometer 204 configured to receive radiation from the system of FIG. 1 as described below. In one non-limiting example, a plurality of optical fibers 205 may be configured to receive laser radiation from one or more laser particles in different directions and transmit the laser radiation to the spectrometer 204. A camera 210 may be included. The system 200 also includes a processor 206. As described below, the process 206 can take any of a variety of forms, including logic gates such as a field programmable gate array (FPGA), or other architectures that process spectra for sorting. The system 200 may further include a memory 208. The memory 208 may be a data acquisition (DAQ) system or card connected to a controller 212 and a switch 214. In some configurations, the processor 206, the memory 208, and the controller 212 are integrated. Further, the system 200 may include a magnetic field generator 216 configured to apply a magnetic field to one or more laser microparticles.

[0028] Referring to FIGS. 1 and 2, the excitation source 202 can be focused upstream 120 of the sorting junction 122 and can record the LP emission signal. During operation, data from the spectrometer 204 can be streamed, for example in real time, and communicated to a processor 206 that processes the spectrum for sorting. The processor 206 first determines whether the intensity value of any pixel exceeds a threshold defined by the camera's noise floor and applies pre-programmed gating criteria to determine "sort (+)" and "non-sort (-)". If the criteria are met, the processor 206 sends the data to the memory 208, and the controller 212 accesses the memory 208 to control the switch 214 to perform the sorting. As will be described later, the switch can be configured to perform acoustic sorting, valve-based routing, inertial separation, or other electromagnetic or physical sorting. As described above, the processor 206, the memory 208, and the controller 212 may be separate components or may be physically or functionally integrated.

[0029] In one non-limiting example, the processor 206 can send transistor-transistor logic (TTL) pulses to a data acquisition (DAQ) card to trigger it to output a pulse train for a predetermined period. For example, the output is a 30 kHz square wave pulse train with a 400 ps period. The flow rate can be set so that the time it takes for the LP to pass through the detection zone is approximately equal to the 30 ps exposure time of the camera 210 or other data acquisition device, and individual LP spectra are read in only one or two frames.

[0030] In one configuration, a square wave pulse train can be supplied to the controller. The controller may be a high voltage amplifier and transmits a 1 kV pk-pk voltage to a pair of microelectrodes 116, 118 operating as switch 214. The voltage may be offset 15 μm from the proximal sidewall of the flow channel 114. Since the aqueous droplet has a higher dielectric constant than the surrounding oil, bound charge can be applied to the aqueous droplet by the electric field. The droplet is then attracted to the electrodes by Coulomb interaction and deflected into the target outflow channel connected to the "sort(+)" outlet 110. This mechanism is known as dielectrophoresis.

[0031] The "non-sort (-)" flow channel may be shorter in the light. In one non-limiting example, the length of the (-) flow path can be made about 5% shorter, resulting in a lower resistive pressure than the sort channel. As a result, when the electrodes 116, 118 remain off, the cells passively flow through the low pressure path and are collected at the "non-sort (-)" outlet 108. In a non-limiting example, the shape of a particular sort junction 122 includes a partition with a gap that occupies a portion of the channel height and functions to further push the droplet towards either the (+) or (-) outlet.

[0032] Figures 3A - 3B show non - limiting examples of the detection of laser emission from laser particles within a microfluidic chip. Figure 3A shows an example of a high - speed video image (14 ps exposure, 10 k frames / second) of a droplet containing cells tagged with a 2 - μm - sized single - semiconductor disk laser flowing through a channel, with a pump light source irradiating a portion of the microfluidic channel. The detection optical system collects the light emitted from the particles as they flow through the readout region. In this case, a distinct signal is observed in several pixels of a line - scan camera of a spectrometer corresponding to a wavelength of 1387 nm (Figure 3B). The recorded linewidth δλ was the full - width at half - maximum (FWHM) of 0.72 nm, limited by the resolution of the spectrometer. Importantly, this narrow linewidth allows the detection system to easily distinguish and resolve multiple laser colors. Figure 3C shows 256 representative recorded spectra from 256 LPs over a bandwidth Δλ of approximately 400 nm, from 1200 nm to 1600 nm, which is the detection range of the spectrometer. Figure 3D shows a histogram of over 460,000 LPs over the entire bandwidth. The distribution itself reflects the recorded diameter of the LPs.

[0033] In this example, although the laser particles are tagged to the cells, the laser particles may not be encapsulated within the cells or attached to the cells. The detection optical system often consists of a spectrometer capable of recording the wavelength of the laser spectrum. However, for certain applications, a photomultiplier tube with a detection bandwidth covering the expected wavelength range of the LP can be used. In this example, a PDMS microfluidic chip on glass was used to construct a 1 cm long straight channel (width × depth = 50 μm × 25 μm). Next, the microfluidic chip was placed on the microscope stage and a 1064 nm pulsed (40 mW, 10 ns pulse width, 2 MHz repetition rate) pump laser was focused at the center of the channel. By introducing a cylindrical lens in front of the objective lens, the focus of the pump laser covered the entire 50 μm channel width and was formed into a line approximately (10 μm × 50 μm). Detection of the spectral signal was performed by collecting light with an emission wavelength of 1100 - 1500 nm and sending it towards a custom-built grating-based spectrometer (resolution 0.5 - 1.0 nm) equipped with a 2048 pixel line scan camera operating at a speed of 29,000 lines per second. Different from most previous studies on spectral flow cytometry (including recent commercial devices), a major advantage of the semiconductor microdisk LP is that only one laser pump source is required. Since the semiconductor LP absorbs light at wavelengths below the bandgap of the particle material, if x and y are selected such that the bandgap of In 1-x Ga x As y P 1-y is less than 1.17 eV (corresponding to a free space wavelength of 1064 nm), laser oscillation is possible. Second, using a high-speed line scan camera detector can significantly increase the sampling of the detected wavelengths compared to the common approach of using multiple optical filters, especially with photomultiplier tubes or avalanche photodiode arrays.

[0034] In a non-limiting example, cells suspended in cell culture medium were injected at one end of a linear channel microfluidic flow device at a flow rate of approximately 40 μL / min. This corresponds to a characteristic linear velocity of approximately 0.5 m / s within the flow channel. Some cells moved at slightly different velocities, but each cell passed through the pump excitation spot in a time equivalent to the exposure time of the camera (30 μs). When an LP-tagged cell passes through the pump excitation, a distinct signal corresponding to a wavelength of approximately 1390 nm in this case is clearly observed in several pixels of the spectrometer's line scan camera. The narrow δλ = 0.72 nm FWHM linewidth is generated by coherent emission from the LP. The peak wavelength at which this laser oscillation occurs is mainly determined by the invariant, i.e., In 1-x Ga x As y P 1-y the material composition and the physical dimensions of the cavity. Thus, the value of the peak wavelength functions as a signature that significantly reduces the complexity of the complete emission profile that is often recorded during spectral flow measurements other than LP-based. Importantly, this narrow linewidth allows the detection system to easily distinguish and resolve multiple laser colors in a single experiment.

[0035] In a non-limiting example, approximately 500,000 LP-tagged cells were flowed through a linear microfluidic channel over a period of 25 minutes. Laser oscillations were observed over a bandwidth of Δλ = 400 nm from 1200 nm to 1600 nm, limited by the detection range of the spectrometer, by using nine different In 1-x Ga x As y P 1-y compositions from (x = 0.26, y = 0.55) to (x = 0.44, y = 0.95). In a simplified theoretical analysis, it is suggested that an LP-based flow cytometer can independently resolve up to approximately Δλ / δλ ≒ 550 colors. Furthermore, analysis of alternative parameters can be performed within the same cytometry channel. For example, fluorescence readouts can also be performed, and morphological characteristics collected by imaging flow cytometry can also be collected. This information can be associated with the spectral signal of the LP.

[0036] Figures 4A-4B show non-limiting examples of binary sorting based on the presence or absence of laser emission from microparticles. Cells are mixed with LP and cultured overnight. Then, approximately 35,000 cells (including 25,000 LP) are removed from the cell culture and suspended in an aqueous cell culture medium. The cell sample is supplied to a sorting device equipped with a gating criterion that transmits a high-voltage pulse when a single laser peak exceeding the noise floor is detected. The cells are flowed into the microfluidic channel via the input channel 400. Optical pumping is provided at the position 410 in front of the sorting switch electrodes 430 and 432. The sorting switch electrodes 430 and 432 may include the microelectrode pairs 116 and 118 of FIG. 1. The flow rate is set so that the time for LP to pass through the detection zone is approximately equal to the 30 μs exposure time of the camera, and the individual LP spectra are read in only one or two frames.

[0037] Droplets containing no laser particles generate spectral data without a distinct laser peak exceeding the detection noise floor of the spectrometer, such as the cell 440 in FIG. 4A. Since there is no laser peak in the spectral data 450 of FIG. 4B, no trigger signal is generated. As a result, the droplet containing it is sent to the "non-sort (-)" collection channel. The non-sort flow channel is 5% shorter than the "sort (+)" flow channel and has a lower resistive pressure than the sort channel. As a result, if the electrodes remain off, the cell passively flows through the low-pressure path and is collected at the (-) outlet.

[0038] On the other hand, in the cell 470 tagged with laser particles in FIG. 4A, a distinct laser peak is generated in the spectral data 480 of FIG. 4B. This triggers the sorting switch, and an electric field applies bound charges to the aqueous droplet, which has a higher dielectric constant than the surrounding oil. Then, the droplet is attracted to the electrode by Coulomb interaction and deflected to the target collection channel connected to the (+) outlet.

[0039] In this non-limiting example of binary LP sorting, cells containing LP are routed separately from barcodeless cells without LP. In this example, HeLa cells were used and mixed with semiconductor microdisk LPs and cultured overnight. The next day, approximately 35,000 cells (including approximately 25,000 LPs) were removed from the cell culture and suspended in cell culture medium. The cell sample was sent to the sorting setup using a gating criterion that sends a high voltage pulse when a single laser peak above the noise floor is detected. The cell samples collected at the (-) and (+) outlets were replated separately and imaged 4 hours after sorting. Figures 5A - 5B show these cells.

[0040] For comparison, prior to taking cell samples from the original culture, the relative distribution of LPs within the cells approximately followed a Poisson distribution (Figure 5C). The binary sort resulted in a significant difference in the relative distribution of LPs within the cells (Figures 5D - 5E). In the (+) outlet sample, 94.1% of the cells contained one or more LPs. In contrast, the (-) outlet sample had few disks, and 99.3% of the cells did not contain disks.

[0041] As shown in Figures 6A - 6D, multiple types of LACS are possible. Figure 6A shows the binary sort demonstrated in the previous example, where cells containing LP are separated from cells without LP. Since LPs can be used to track mammalian cells in studies of tumor invasion and migration, such sorting operations can be used to isolate subpopulations of cells that can be tracked. Alternatively, the LP tag has been proposed as a single-cell sequencing probe, and its luminescence signature can be used to track individual cells through an RNA sequencing workflow and link knowledge of each cell's spatial location to information about its gene expression. In some single-cell sequencing protocols, cells are typically sorted prior to sequencing by conventional FACS to target subpopulations for sequencing. Similarly, this type of LACS sorter can enrich cells suitable for sequencing by selecting tagged cells.

[0042] Figure 6B shows a sort of the same type, where cells containing a predetermined number of LPs are separated from cells that do not contain that number of LPs. For example, the (+) outlet contains cells with 3 LPs, and the (-) outlet contains cells without 3 LPs. Importantly, since each cell at the (+) outlet is tagged with 3 LPs, each cell contains a unique spectral barcode that cannot be achieved with fewer LPs. Further, it may be desirable to remove cells where there are too many LPs that may adversely affect natural operation.

[0043] Figure 6C shows a wavelength-based LP sort. This allows separation of a population of LPs or LP-tagged cells with very specific emission spectra. Using this, a monoclonal batch of high-purity spectral phosphors with emission bandwidths much narrower than those of fluorescent phosphors can be created. For example, using an LP sorting system, LPs or LP-tagged cells with emission characteristics over a nanometer spectral range can be separated, in contrast to the tens-of-nanometers FWHM emission characteristics of fluorescent phosphors. Or, the sorting function can be used to remove particles with poor optical performance (such as broad spectral linewidths or low intensity).

[0044] Figure 6D shows a sort using "multiplex" laser particles. A multiplex is an LP that emits laser light at multiple laser wavelengths and can exhibit thousands or more unique spectral barcodes. Specific multiplexes can be sorted and separated based on spectral data. Thus, the user can specifically determine which cells to separate based on the unique multiplex barcode of each cell. Such a sorter can map sorting decisions using a pre-computed lookup table. Thus, when a particular spectrum is confirmed by a spectrometer, only a query within the table is needed to make a sorting decision.

[0045] Figures 7A - 7D show non - limiting examples of sorting based on spectral barcodes. First, a wavelength sorting gating strategy for short - wavelength pass or long - wavelength pass was tested. The FPGA linked to the spectrometer was programmed to identify the pixel of the highest intensity in real - time each time a laser event was recorded. If this pixel corresponded to a wavelength within the sorting window, a high - voltage electric field was applied to the electrodes, and the passing droplets were deflected towards the appropriate outflow channel. When multiple laser peaks were emitted from a droplet (presumably because there were two cells or a single cell with two LPs in the droplet), the gating conditions were set such that a sorting signal was emitted only if all the emitted peaks were within the sorting window. In the experiment, short - wavelength gating conditions with a cut - off wavelength of 1450 nm were used. In Figure 7A, cell 720 contains an LP whose emission spectrum 740 (Figure 7B) is outside the gating window. Thus, cell 720 is sorted to the (-) outlet. In contrast, cell 760 in Figure 7C contains an LP whose emission spectrum 780 (Figure 7D) is within the gating window. Thus, cell 760 is sorted to the (+) outlet.

[0046] Figures 8A - D show non - limiting examples of band - pass sorting. As a selection criterion, a sorting window with a width of 5 nm centered at 1285 nm was applied. LPs obtained from a single semiconductor wafer using In 0.75 Ga 0.25 As 0.54 P 0.46 were used, which cover a spectral range of 1245 - 1340 nm. The LPs were co - cultured with HeLa cells and flowed through the sorting chip. The cells collected from each outlet were re - plated into a culture well plate, and the LP emission from the cells was measured using a confocal microscope modified previously to employ a high - resolution spectrometer. In this non - limiting example, the cells collected from the (+) outlet contained LPs that mainly emitted lasers at wavelengths within a defined 5 - nm window close to 1285 nm. The routing accuracy was consistent with that obtained with the wavelength - independent binary sorting described above.

[0047] In this way, a subpopulation of cells associated with a specific tag color can be actively enriched. As a result, this leads to the separation of cells marked with emitters of unparalleled spectral purity. In the above example, In 0.75 Ga 0.25 As 0.54 P 0.46 LP and HeLa cells were used. The fluorescence emission of each LP obtained by measuring the spectrum below the pump threshold (-10 pj) had a FWHM of 70 nm and was centered at approximately 1275 nm. This low spectral purity leads to significant crosstalk that hinders multiplexed multi-fluorescence measurements. Similarly, the spectral distribution of -100 nm of the laser wavelength is large. In the large-scale manufacture of microdisk LPs by photolithography, inevitably, due to the non-uniformity of the fine processing across the entire surface of the original semiconductor wafer, a nanometer-scale variation occurs in the cavity diameter. When the diameter increases by 1 nm, the emission wavelength changes by approximately 1 nm within a specific optical mode. Due to the influence of the non-uniformity introduced during LP manufacturing, the emission wavelengths of the LPs within the set are substantially randomized.

[0048] A sorting window of ±2.5 nm was used as reference data that functions as a selection criterion or gating strategy. The FPGA linked to the spectrometer was programmed to identify the pixel with the highest intensity in real time every time a laser event was recorded. If this pixel corresponded to a wavelength within the sorting window, the electric field was used again, and the passing droplets were deflected towards the appropriate outflow channel. If it was determined that multiple laser peaks were emitted from the same droplet, the droplet was actively deflected only if both emission peaks were laser-irradiated within the sorting window.

[0049] After detecting the laser emission that occurred within the sorting window 800 times, the cells were re-plated 4 hours later. The LP emission from the cells collected from each outlet was measured using a microscope equipped with a commercially available spectrometer. From the histogram of the recorded peak laser emission wavelengths, it was confirmed that most of the LPs emit at approximately 1285 nm. In contrast, measurements of the LPs collected from other outflow channels showed that these cells are tagged with LPs that have little LP near 1285 nm and emit laser light over a much broader 100 nm spectral bandwidth.

[0050] The emission intensity of a specific type of LP, such as a microdisk-shaped LP, can vary significantly depending on the angle. In the case of a microdisk LP, most of its laser emission is within the plane of the disk. Therefore, when the orientation of the disk is flat within the microchannel (i.e., when the disk plane is orthogonal to the optical axis of the collection optics), detection can be difficult. Several methods can be considered to overcome this problem. First, multi-axis collection optics may be used. In one embodiment, two or more collection objective lenses arranged at different angles may be used. In another embodiment, two or more optical fiber collectors may be used after a single objective lens. FIG. 9A shows such a collection system with two optical fibers that collect light at two different angles. To collect the output emission from the laser particles, three or more optical fibers may be used. These fibers are connected to the spectrometer using appropriate coupling optics. This arrangement can reliably detect the laser emission from arbitrarily oriented microparticles, minimizing the intensity change within the predetermined detection dynamic range of the spectrometer and maximizing the probability of detecting the laser peak.

[0051] Figure 9B shows an alternative technique. Magnetic microdisks have been demonstrated, for example, using thin layers of magnetic materials such as permalloy. These disks exhibit a magnetic dipole moment, and when an external magnetic field is applied, the orientation of the disk changes and its moment aligns with the external magnetic field. This effect has been used to demonstrate the manipulation of microdisk objects within a fluid. Such magnetic layers can be incorporated into the LP and the external magnetic field and applied so that the disks align with the optical axis of the collection system as they flow.

[0052] Yet another technique involves incorporating a highly scattering medium into the flowing fluid. Liquids that support high levels of scattering include liquids containing T i O2 or Intralipid. Typically, when the concentration of Intralipid is about 1%, it approaches the scattering characteristics of brain tissue. By surrounding the flowing disk with a scattering solution, light can be redirected more uniformly over 4π steradians and redirected to the collection objective lens. To prevent unwanted interactions between the scatterer and the optical modes of the LP, a coating such as silica can be applied to surround the bare LP. Such a scattering fluid may also reduce the ability of the pump to penetrate the LP. However, since the peak signal of LP emission saturates quickly, more powerful pumping may simply result in a broadening of the linewidth and may not contribute to an improvement in the signal-to-noise ratio in many detection systems. Therefore, at high flow rates, the effect on collection efficiency may become dominant.

[0053] Another approach is to use omnidirectional laser particles designed to minimize angle-dependent radiation. Various methods for minimizing angle dependence are known in the art, including coating the laser particles with scattering nanoparticles.

[0054] In the example, microdisk laser particles are used, but any type of laser particle known in the art, such as spherical bead microparticles or nanowires, can be used.

[0055] Droplet microfluidics provides a convenient technology platform that can be used to manipulate individual micro-objects. These objects are encapsulated using aqueous micro-droplets surrounded by immiscible oil and confined within the droplets so that they can be manipulated individually. Recently, a commercially available droplet fluid sorter (Onchip Biotechnologies) that can encapsulate and sort fluorescent objects as small as individual bacteria has been released. Droplet-based sorting can also be used to manipulate individual LPs or cells.

[0056] Figure 10A shows an exemplary utility for generating droplets each containing only one cell or one LP. For example, droplet 1000 encapsulating cell 1010 is fused with droplet 1020 containing LP 1030 to efficiently combine these two into one droplet 1040. This utility can avoid the limitations of random tagging affected by Poisson statistics.

[0057] Figure 10B shows another utility pairing of two LPs (1050 and 1052), generating droplet 1060 containing two LPs each. With further operations, droplets containing any specific number of LPs can be generated. When a droplet containing the required number of LPs is generated, the exact number of LPs is supplied to the cells by further fusion with droplets containing cells. Figure 10C shows two droplets containing single microdisk InGaAsP LPs 1070 and 1072.

[0058] Another exemplary utility is suitable for the formation of multiplex-type laser particles. The captured LPs are then attached by chemical means within the droplet. Such an operation can controllably form a multiplex with a specific number of LPs. For example, a triplet (a multiplex composed of three individual laser microparticles) can be formed by fusing a droplet with a pair of LPs and a droplet with only one LP. A quartet multiplex can be formed by combining two droplets each containing two LPs.

[0059] The fusion of a pair of droplets can be performed using previously developed droplet fusion devices. This device confines a "preceding" droplet between narrow openings, allowing a "subsequent" droplet to collide with it. FIG. 10D shows the experimental results of fusing two droplets 1080 and 1082 at consecutive time points before, during, and after fusion, resulting in the formation of a fused droplet 1090. Since the microparticles remain encapsulated within the individual droplets, they can maintain a stable state for several hours during which the droplets, and thus their contents, can be manipulated.

[0060] Although a two-way sorting architecture with two collection channels has been described, these embodiments can be extended to two or more collection channels. In state-of-the-art commercial flow sorters, four-way or six-way sorters are commonly used. Numerous scenarios for sorting cells and laser particles are envisioned. For example, laser particles with a first spectral barcode may be routed to a first collection channel, and laser particles with a second barcode may be routed to a second collection channel, and so on.

[0061] Another embodiment of the present invention is a sorter that uses a recirculator. One application of this embodiment is to split a multi-color LP set into multiple monochromatic batches. For example, the starting set of LPs spans a 100 nm spectral window. If they are sorted based on a spectral window of, for example, ±2.5 nm, 20 monochromatic batches can be created from a single 100 nm spectral window. Using a smaller window, such as 1 nm, more batches with the same laser wavelength within the 1 nm window can be created.

[0062] Figures 11A - 11D show the procedure of a recirculation platform for efficiently performing this operation by recirculating the contents included in one of the collection channels or outlets. Dielectrophoretic droplet sorters are one possible sorting mechanism that can be applied, but alternative sorting mechanisms such as acoustic sorting, valve - based routing, inertial separation, and other techniques known in the art can also be used. Droplet sorters have the advantage of preventing sedimentation and aggregation between LPs within individual droplets. Figure 11A shows the first step of automatically generating a monochromatic batch from a relatively polychromatic LP set. First, the LP suspension is injected into and collected by a droplet - generating chip. In the second step, the fluid switch is toggled (toggle switch 1) to enable the droplet - formed LPs to be moved to another vial (Figure 11B). Yet another toggle switch injects the LPs into the sorting device and separates the monochromatic batch of LPs based on a reference spectral window specified by the user (Figure 11C). The unsorted LPs are collected again. Finally, fluid switch 2 is toggled and the LPs are reinjected into the sorting device. When switch 3 is toggled, different LP batches can be collected based on different spectral collection windows (Figure 11D). LPs that are not part of the new monochromatic batch are collected again. Then, the cycle is repeated from the steps shown in Figure 11B until all monochromatic batches are formed.

[0063] The above - mentioned laser - particle - based sorting method can be combined with conventional fluorescence flow cytometry techniques. Since LPs emitting at infrared wavelengths do not show crosstalk with common phosphors, LPs are compatible with established fluorescence - based techniques.

[0064] LP tags have been proposed as single-cell sequencing probes, which can track individual cells through the RNA sequencing workflow by luminescence signatures and associate knowledge about the spatial position of each cell with information about its gene expression. In some single-cell sequencing protocols, it is common to sort cells by conventional FACS before sequencing to target subpopulations for sequencing. Similarly, in binary LACS, cells suitable for sequencing can be enriched by selecting tagged cells. Wavelength-based LP sorting is expected to be more versatile. For example, specific cells identified in imaging experiments can be separated for further analysis.

[0065] LACS can be used to separate cells based on previously identified LP barcodes by techniques such as microscopy and flow cytometry. These techniques can associate phenotypic characteristics such as protein-tagged fluorescence expression with individual cells. By reading the LP signature at each step of the workflow, a comprehensive cell profile can be constructed and separated as needed.

[0066] This technology can be used to sort LPs regardless of the presence or absence of cells in droplets, in addition to cell sorting. Currently, LPs are manufactured such that each particle has a random wavelength and is uniformly distributed across the gain bandwidth of the semiconductor. However, for some experiments, it would be convenient to be able to generate batches of LPs with the same wavelength, similar to the case of phosphors with the same fluorescence spectrum. By repeating wavelength-selective sorting, batches of LPs of different colors can be created and used for various applications such as highly multiplexed cell type labeling for in vivo imaging.

[0067] Figure 12 shows an example of batch laser particles having substantially the same barcode characteristics. A total of 11 batches from LP groups 1200 to 1220 are shown. For simplicity, only three particles per batch or group are shown along with their emission spectra. Each particle is assumed to emit a single laser peak. In this figure, the laser peak wavelength of all particles in the first group is 700 + / - 3 nm. The second group is characterized by a peak wavelength of 710 + / - 3 nm. The barcode characteristics are defined by the wavelength range of each group. All 11 groups have clearly different barcode characteristics. The separation or window of the wavelengths may be less than + / - 1 nm.

[0068] Within each batch composed of many particles, two or more laser particles may have approximately the same wavelength, making them difficult to distinguish. As long as particles from different groups can be distinguished, this is not a problem in many applications.

[0069] In the following examples, the manufacturing methods of LPs and microfluidic devices and their usage in the above examples will be described in more detail. <Example>

[0070] Materials and Methods

[0071] Manufacture and Transfer of LPs

[0072] For the manufacture of LPs, an epitaxially grown wafer was used. The wafer was based on an InP substrate and had an undoped InP buffer layer 300 nm thick. On top of this layer, one or more InGaAsP layers 200 nm thick were grown at 300 nm intervals. A 2 μm thick SU8-2002 (MicroChem) photoresist was applied to the wafer surface, followed by soft baking at 65 °C for 1 minute and 95 °C for 2 minutes. The wafer was exposed using an i-line and h-line mercury arc lamp at 60 mJ / cm 2Exposed through chromium on a quartz mask composed of an array of hexagonal circles at a dose of (Karl Suss MJB4 mask aligner). After that, it was baked at 65 °C and 95 °C for 1 minute and developed with SU8 developer (MicroChem). Next, a hard bake was performed on a hot plate at 190 °C for 10 minutes, and then the wafer was descumed using O2 plasma (Anatech Barrel SCE 160). Next, the remaining circular photoresist features were used as a mask, and the columns were dry etched onto the wafer using chlorine chemistry (Oxford Instruments PlasmaPro 100 Cobra 300). Next, a cleaning sequence consisting of a 3-minute plasma cleaning using CF4 and O2 (Oxford Instruments PlasmaPro 100 Cobra 300), a 30-second immersion in 1:1 H2SO4:H2O, and a final O2 plasma cleaning (Matrix 105) was used to remove the remaining SU8. Next, the microdisk LPs were detached from the substrate by dissolving the InP support in 3:1 HCl:H2O for 30 seconds. Next, the LPs were washed at least three times with deionized water by repeating centrifugation and removal of the supernatant. Next, the LPs suspended in water were added to the cells, and an appropriate amount of 10×PBS was added to maintain an isotonic solution.

[0073] Fabrication of Microfluidic Devices Polydimethylsiloxane (PDMS) microfluidic devices were created using an SU8-on-silicon master mold. To create this mold, a two-layer device mold was created. The first layer consisted of channels that separated the sort dividers and sort outflow channels that allowed for pressure equalization. To create this layer, SU8-3010 (Kayaku Advanced Materials) was spin-coated onto a Si wafer (University Wafer) at 3000 rpm and soft baked at 95 °C for 5 minutes. Next, the first layer of the pattern was exposed at 2500 mJ / cm 2Exposure was carried out using a 365 nm wavelength laser writer (Heidelberg Instruments MLA150). Next, post-exposure baking was performed at 95 °C for 5 minutes, and then development was carried out with propylene glycol methyl ether acetate for 5 minutes. The second layer defined the microfluidic flow channels and electrode channels. To form this layer, the process was repeated using SU8-3025. Using the laser writer, patterns were written aligned with the first layer. Next, 1 drop of 1H, 1H, 2H, 2H-perfluorooctyltrichlorosilane (Oakwood Chemical) was placed into the master mold in a vacuum chamber for 1 hour, and then baked at 180 °C for 1 minute. This prevented the PDMS from sticking firmly to the mold. Next, a 10:1 mixture of base elastomer and curing agent PDMS (Sylgard 184 Silicone Elastomer Kit) was poured into the mold and cured at 65 °C for 2 hours. Thereafter, the molded PDMS was peeled off the wafer. Next, using a 1.2 mm diameter biopsy punch (Harris Uni-Core), holes were made in the PDMS for each of the three inlets, two outlets, and four electrode openings. Next, the PDMS was bonded to the glass slide substrate using O2 plasma bonding (Plasma Etch) and baked at 70 °C for 2 hours to further strengthen the bond. 1 / 16″×0.006″ OD×ID PTFE tubes (Valeo Instruments) were connected to the inlets and outlets of each flow channel. A pressure pump (Fluigent Flow EZ) and an in-line flow meter (Fluigent Flow Unit) were connected to the other end of the tube linked to the inlet. Aquapel (Pittsburgh Glass Works) was flowed into the device to make the glass surface hydrophobic. After leaving it for 30 seconds, the device was rinsed with FC-40 and dried using an N2 air gun. To create the electrodes, the microfluidic device was placed on a hot plate at 80 °C, and low melting point In 51 Bi 32.5 Sn 16.5Indium Corporation Indalloy solder was supplied to one of the openings of the energized electrode channel and one of the openings of the ground electrode channel. While on the hot plate, two Digi-Key 87224-1 gold connectors were placed at each of the two openings. Loctite 352, a UV-curing adhesive, was used to strengthen the connection between the PDMS and the gold connectors.

[0074] Cell culture

[0075] HeLa cells (ATCC) were cultured in a solution consisting of Dulbecco's modified Eagle medium, 10% (v / v) fetal bovine serum (FBS), and 1% (v / v) penicillin-streptomycin. In the LP imaging experiment, to ensure isotonicity, the required amount of LP was added directly to the cell culture together with 10× PBS. After 24 hours, to remove the cells in which LP had internalized from the plate, the culture was first washed three times with PBS and incubated for 5 minutes with TiypLE Express Enzyme (Fisher Scientific) (Thermo Scientific Heracell 240i). Next, the floating cells were removed from the culture plate and passed through a 40-μm strainer (MTC Bio SureStrain Premium Cell Strainers) to remove cell clumps. Next, to prevent individual cells from adhering to the mesh, excess medium was passed through the strainer. Next, centrifugation was used to pellet the cells and the supernatant was aspirated. Next, the cells were resuspended in a solution of 79% cell culture medium, 17% OptiPrep (Sigma Aldrich), 1.28% Dnase I (Thermo Scientific), and 2.55% Dnase reaction buffer (Thermo Scientific). OptiPrep prevented the cells from precipitating during the flow experiment, and Dnase minimized cell aggregation due to free DNA strands. After the sorting experiment, approximately twice the droplet volume of Pico-Break 1 (Sphere Fluidics) was added to the collection tube, and the mixture was placed on ice for 1 minute to recover the cells from the droplets surrounded by oil. Next, the upper solution was gently pipetted up and down. This dissolved the droplets, leaving a two-layer liquid of oil at the bottom and cell culture medium containing cells at the top. Next, an excess amount of cell culture medium containing 20% OptiPrep was added. Next, the collection tube was centrifuged at 100 g for 5 seconds to make the interface between the oil and the aqueous suspension of cells clearer. By adding OptiPrep, the possibility of the cells adhering to the oil-aqueous interface was minimized. Next, the upper layer was carefully removed and transferred to a clean centrifuge tube with a pipette.Next, an OptiPrep-free cell culture medium was added in excess to this tube to dilute the OptiPrep, and the cells were pelleted by centrifugation. After pellet formation, the supernatant was aspirated and replaced with fresh medium. Then, the cells were re-plated into wells with a glass bottom for further imaging and characterization.

[0076] Cell sorting experiment

[0077] A microfluidic device was placed on a microscope stage. The microscope was equipped with a high-speed camera (Integrated Design Tools M3), a pump source for exciting the LP, and a custom-built spectrometer for measuring the laser output. The pump laser focus was placed next to the sorting junction. Of the three input flow channels, one was connected to an aqueous reservoir composed of LP-tagged cells. The other two were filled with droplet generation oil (Bio-Rad QX200 Droplet Generation Oil). The flow was driven using a pressurized source (Fluigent Flow EZ), and the velocity was controlled by a flow meter (Fluigent Flow Unit) with closed-loop feedback. The two outflow channels were each connected to a collection tube. First, the flow rate to the spacer inlet was set to 35 μL / min, the flow rate to the oil generation inlet was set to 5 μL / min, and the flow rate of the aqueous cell solution was set to 2 μL / min. As a result, droplets containing tagged cells were generated. Next, these droplets were spaced apart by the oil from the spacer inlet, enabling sorting of one droplet at a time. Next, the flow pattern of the droplets was monitored using the high-speed camera, and the flow rate was adjusted to achieve the correct timing. Next, one end of the live electrode and one end of the ground electrode were connected to a high-voltage amplifier (TREK Model 2210-CE) with alligator clips. The amplifier amplified a square wave train emitted from a DAQ card (National Instruments PCIe-6321) triggered by a pulse transmitted from an FPGA (National Instruments PCIe-1473R) when appropriate sorting conditions were met. Optical measurements of the LP emission during flow were collected by pumping the LP using a 10 ns pulse width at 1064 nm with a repetition rate of 2 MHz (CNILaser FL-1064-Nano-LAB). The excitation light was focused using a 10×0.3NA objective lens (Leica HC PL FLUOTAR). The emitted light was collected and dispersed using a custom-built grating spectrometer equipped with a line scan camera (Sensors Unlimited 2048 L).This camera was connected to an FPGA (National Instruments PCIe-1473R) that determines whether the sorting criteria are met. The sorting accuracy was evaluated by comparing the input and output of the flow device using a confocal microscope (Olympus Fluoview 3000). This microscope was capable of acquiring bright-field images of cells and LPs and was equipped with a cell incubator (Tokai Hit). Furthermore, the microscope was modified to measure LPs using a 1060-1070 pump laser (Spectra Physics VGEN-ISP-POD) operating at 2 MHz with a pulse width of 10 ns. Spectra were collected using a 20×0.45NA objective lens (Olympus IMS LCPLN20XIR) and transmitted to a spectrometer (Andor Kymera 328i) equipped with a line scan camera (Sensors Unlimited 2048 L). All camera data were saved and analyzed using custom code based on MATLAB® and Python.

[0078] The present invention has been described with respect to exemplary embodiments, and it should be understood that many equivalents, alternatives, modifications, additions, and revisions can be made in addition to those explicitly stated, as well as in various ways of combining different features of the foregoing versions, and they are within the scope of the present invention. The above detailed description has shown, described, and pointed out novel features applicable to various embodiments, but it will be understood that various omissions, substitutions, and changes in the form and details of the illustrated devices or algorithms can be made without departing from the spirit of the disclosure. As will be recognized, certain embodiments of the disclosure described herein can be implemented in a form that does not provide all of the features and advantages described herein because some features can be used or implemented separately from other features. The scope of the specific disclosure disclosed herein is indicated by the appended claims rather than the foregoing description. All changes within the meaning and scope of equivalence of the claims are to be included within their scope.

Claims

1. A sample loader configured to receive a sample containing one or more laser microparticles, each laser microparticle being configured to generate laser radiation having one or more different spectral peaks when excited, the sample loader, A spectrometer that receives the laser radiation from the one or more laser microparticles and generates spectral data, A processor configured to receive the spectral data and generate a sorting signal, A switch that receives the sorting signal and is configured to route the one or more microparticles to a specific one of a plurality of collection channels based on the sorting signal, A flow sorter comprising.

2. Further comprising a database containing reference data, The sorting signal is generated based on a comparison between the reference data in the database and the spectral data generated by the spectrometer, The flow sorter according to claim 1.

3. The flow sorter according to claim 1, further comprising an excitation source and a photodetector configured to detect fluorescence from the sample.

4. The flow sorter according to claim 1, wherein the one or more laser microparticles are bound to cell entities.

5. The flow sorter according to claim 1, wherein the one or more laser microparticles in the sample are encapsulated in at least one emulsion droplet.

6. The switch includes one of a mechanical, electrostatic, or valve actuator, The flow sorter according to claim 1.

7. The spectrometer has an optical resolution of less than 1 nm, The flow sorter according to claim 1.

8. The one or more laser microparticles are composed of a semiconductor disk laser, The flow sorter according to claim 1.

9. Further comprising a magnetic field generator configured to apply a magnetic field to the one or more laser microparticles, The flow sorter according to claim 1.

10. Further comprising a plurality of optical fibers configured to receive the laser radiation from the one or more laser microparticles from different directions and transmit the laser radiation to the spectrometer, The flow sorter according to claim 1.

11. The processor is configured to control the switch to route microparticles having substantially similar barcode characteristics in the spectral data to a common one of the collection channels, The flow sorter according to claim 1.

12. The barcode characteristics include at least one of the wavelength and number of times of the laser radiation, The flow sorter according to claim 11.

13. The processor is configured to control the switch to route laser microparticles having substantially the same laser peak wavelength to a common one of the collection channels, The flow sorter according to claim 11.

14. The processor is configured to control the switch to route laser microparticles having one or more substantially different barcode characteristics in the spectral data compared to the one or more laser microparticles routed to different collection channels among the plurality of collection channels, The flow sorter according to claim 1.

15. Further comprising a recirculator configured to supply the one or more laser microparticles routed to the sample loader. The flow sorter according to claim 1.

16. Loading a sample containing one or more laser microparticles into a microfluidic system; Exciting the one or more laser microparticles to generate laser emission for each of the one or more laser microparticles; Analyzing the spectrum of the laser emission to determine the characteristics of the laser emission of each of the one or more laser microparticles; Sorting the one or more laser microparticles based on the characteristics as each of the one or more laser microparticles travels through the microfluidic system, and sending each of the one or more laser microparticles having predetermined spectral characteristics to a common outlet of the microfluidic system; A method for sorting laser microparticles comprising.

17. Each of the one or more laser microparticles is bound to a cell. The method according to claim 16.

18. The one or more laser microparticles in the sample are encapsulated in at least one emulsion droplet. The method according to claim 16.

19. The characteristics include a laser peak wavelength. The method according to claim 16.

20. Further comprising applying a magnetic field to each of the one or more laser microparticles traveling through at least a portion of the microfluidic system. The method according to claim 16.

Images