Systems and methods for identifying cells using hyperspectral imaging and other techniques

Abstract

Some aspects of the present disclosure are generally related to systems or methods for identifying cells using hyperspectral imaging or other imaging techniques. According to some embodiments, the system may be configured to perform hyperspectral imaging on a single cell, e.g., a single bacteria cell. In some embodiments, the systems described herein may include a microfluidic device having a plurality of cell growth trenches configured to separate and / or grow cells therein. In some such embodiments, the microfluidic device may facilitate high throughput separations and / or identification of cells from a sample, e.g., by acquiring hyperspectral profiles of cells separated and / or grown in the microfluidic device. Still other aspects of the present disclosure are generally directed to methods of using the systems, kits containing the system, or the like.

Description

[0001] SYSTEMS AND METHODS FOR IDENTIFYING CELLS USING HYPERSPECTRAL IMAGING AND OTHER TECHNIQUES

[0002] RELATED APPLICATIONS

[0003] This application claims the benefit of U.S. Provisional Patent Application Serial No. 63 / 646,534, filed May 13, 2024, entitled “Systems and Methods for Identifying Cells Using Hyperspectral Imaging and Other Techniques,” by Paulsson, et al., U.S. Provisional Patent Application Serial No. 63 / 646,582, filed May 13, 2024, entitled “Identifying Bacteria and other Organisms by Determining Growth Profiles,” by Paulsson, et al., and U.S. Provisional Patent Application Serial No. 63 / 646,599, filed May 13, 2024, entitled “Microscopy Systems and Methods for Identifying Bacteria and Other Organisms in Microfluidic Devices,” by Paulsson, et al. Each of these is incorporated herein by reference in its entirety.

[0004] GOVERNMENT FUNDING

[0005] This invention was made with government support under AY2AX000005-01 awarded by Advanced Research Projects Agency for Health (ARPA-H) and under W91 INF-19-2-0018 awarded by U.S. Army Research Office (ARO). The government has certain rights in this invention.

[0006] FIELD

[0007] Certain aspects of the present disclosure are generally directed to systems and methods for identifying cells using hyperspectral imaging or other imaging techniques, e.g., in microfluidic devices.

[0008] BACKGROUND

[0009] Cell screenings play a fundamental role in biology and make it possible to identify one or more cells of interest based on a desired phenotype and / or genotype associated with the target cells. Current cell screening techniques only provide endpoint low-resolution snapshots, and offer low throughput when identifying multiple types of cells from within a sample. In some cases, current techniques are often limited by scalability, screening errors due to phenotypic mischaracterization, long time to identification, etc. Thus, more effective systems and methods for cell screening are needed.

[0010] SUMMARY Certain aspects of the present disclosure are generally directed to systems and methods for identifying cells using hyperspectral imaging or other imaging techniques, e.g., in microfluidic devices. The subject matter of the present disclosure involves, in some cases, interrelated products, alternative solutions to a particular problem, and / or a plurality of different uses of one or more systems and / or articles.

[0011] Some aspects are related to methods. In some embodiments, the method comprises acquiring hyperspectral profiles of at least some of a plurality of cells within a plurality of cell growth trenches of a microfluidic device.

[0012] In some embodiments, the method comprises growing a plurality of cells within a plurality of cell growth trenches under a first set of conditions; acquiring a first hyperspectral profile of at least one of the cells within at least one of the cell growth trenches; thereafter, growing the plurality of cells within the plurality of cell growth trenches under a second set of conditions; and acquiring a second hyperspectral profile of the at least one of the cells within the at least one of the cell growth trenches.

[0013] In some embodiments, the method comprises acquiring a hyperspectral profile of only a single cell.

[0014] In some embodiments, the method comprises containing a plurality of cells from a sample of one or more different unknown species of unicellular organisms within a microfluidic device; and identifying a species of the unicellular organisms within the microfluidic device in less than or equal to 1 hour after containing the cells within the microfluidic device.

[0015] Some aspects are related to systems. In some embodiments, a system comprises a microfluidic device, comprising: a cell flow layer comprising a growth channel having an inlet portion, an outlet portion, an inlet valve portion associated with the inlet portion, and an outlet valve portion associated with the outlet portion; a plurality of cell growth trenches configured to contain cells, the plurality of cell growth trenches being fluidically coupled to the growth channel; and a control layer configured to control flow of the fluid in the cell flow layer; and a light source configured for fluorescence microscopy and positioned to direct light at at least a portion of the plurality of cell growth trenches of the microfluidic device; a mask positioned along an optical path between the light source and the device; a detector positioned to receive fluorescence from the cells illuminated by the light; and a processor configured and arranged to perform hyperspectral analysis on the fluorescence received by the detector.

[0016] Other advantages and novel features of the present disclosure will become apparent from the following detailed description of various non-limiting embodiments of the disclosure when considered in conjunction with the accompanying figures.

[0017] BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Non-limiting embodiments of the present disclosure will be described by way of example with reference to the accompanying figures, which are schematic and are not intended to be drawn to scale. In the figures, each identical or nearly identical component illustrated is typically represented by a single numeral. For purposes of clarity, not every component is labeled in every figure, nor is every component of each embodiment of the disclosure shown where illustration is not necessary to allow those of ordinary skill in the ail to understand the disclosure. In the figures:

[0019] FIGS. 1A-1B arc schematic representations of a system, according to some embodiments;

[0020] FIG. 2A is a schematic representation of a top view of one embodiment of a microfluidic device, according to some embodiments;

[0021] FIG. 2B is a schematic representation of a perspective view of the microfluidic device of FIG. 2A, according to some embodiments;

[0022] FIG. 2C is a schematic representation of a partially exploded view of the microfluidic device of FIG. 2B, according to some embodiments;

[0023] FIG. 3A is a schematic representation of a top view of a section of the microfluidic device of FIG. 2A in an unactuated state, according to some embodiments;

[0024] FIG. 3B is a schematic representation of a top view of a section of the microfluidic device of FIG. 2A in an actuated state, according to some embodiments;

[0025] FIG. 4A is a schematic representation illustrating the step of cell injection into the growth channel of FIG. 3A, according to some embodiments;

[0026] FIG. 4B is a schematic representation illustrating the step of flushing the growth channel of FIG. 3A, according to some embodiments;

[0027] FIG. 5 is a conceptual view of a hyperspectral hypercube of images of an acquired hyperspectral profile, according to some embodiments; FIG. 6 shows images collected from a cell as a function of excitation wavelength and to determine morphology, as well as a plot of the projection of the spectral signal as a function of excitation wavelength, according to some embodiments;

[0028] FIG. 7 is a confusion matrix applied in the context of 25 distinct bacterial populations, according to some embodiments;

[0029] FIG. 8 is a confusion matrix applied in the context of 12 bacterial strains, according to some embodiments;

[0030] FIG. 9 shows the autofluorescence signatures of three different bacterial populations, according to some embodiments;

[0031] FIG. 10 shows the accuracy of a model for identifying cells on cells in a stationary phase as a function of time along a growth curve, according to some embodiments;

[0032] FIG. 11 is a confusion matrix applied in the context of 35 distinct bacterial populations, according to some embodiments;

[0033] FIG. 12 is a confusion matrix applied in the context of 30 distinct bacterial populations, according to some embodiments;

[0034] FIGS 13A-13C schematic illustrations and images of collecting hyperspectral profiles from bacteria in microfluidic devices having a plurality of cell growth trenches, according to some embodiments;

[0035] FIGS. 14A-14B are hyperspectral profiles and a corresponding confusion matrix, respectively, collected for eight bacterial species, according to some embodiments;

[0036] FIGS. 15A-15B show confusion matrices for various bacterial species, according to some embodiments;

[0037] FIGS 16A-16H are plots of autofluorescence measurements in the presence and absence of 5 -ALA, according to some embodiments;

[0038] FIG. 17 are images of cells grown in different media, according to some embodiments;

[0039] FIGS. 18A-20C show schematic diagrams and images related to improvements in experimental configurations for hyperspectral imaging, according to some embodiments;

[0040] FIGS. 21A-22D are raw data and confusion matrices for identifying bacterial species in different growth media, according to some embodiments; FIGS. 23A-25 show data related to hyperspectral imaging of bacterial cells before and after treatment with antibiotics, according to some embodiments.

[0041] DETAILED DESCRIPTION

[0042] Some aspects of the present disclosure are generally related to systems or methods for identifying cells using hyperspectral imaging or other imaging techniques. According to some embodiments, the system may be configured to perform hyperspectral imaging on a single cell, e.g., a single bacteria cell. In some embodiments, the systems described herein may include a microfluidic device having a plurality of cell growth trenches configured to separate and / or grow cells therein. In some such embodiments, the microfluidic device may facilitate high throughput separations and / or identification of cells from a sample, e.g., by acquiring hyperspectral profiles of cells separated and / or grown in the microfluidic device. Still other aspects of the present disclosure are generally directed to methods of using the systems, kits containing the system, or the like.

[0043] Cell screening plays a fundamental role in biology, and makes it possible to identify one or more cells of interest based on a phenotype and / or genotype. Cell screening may be desirable in any of a variety of applications, for example, determining an illness in a human or other organism, in determining food spoilage, and / or determining contamination in the environment. Various cell screening methods have been developed over time. For example, a conventional technique for genotyping cells is sequencing, where at least a portion of the genome of a cell is amplified and sequenced to identify the cell. Phenotypic identification typically may also include one or more of any of a variety of tests, including biochemical reactions, serologic tests, growth requirements, gram stain, bacterial morphology, and information about from where the cell was isolated. In either case of genotypic or phenotypic cell screening, there are a variety of challenges, including low screening throughput, slow identification times, limited potential for scalability and automation, limited screening capabilities for complex cell populations, and phenotypic mischaracterizations caused by screening errors.

[0044] Accordingly, some aspects of the present disclosure are generally related to systems and methods for improved cell screening, e.g., by using hyperspectral imaging, or other imaging techniques. Advantageously, in some embodiments, the methods described herein may identify a cell with relatively high accuracy. The identification may stem from acquiring hyperspectral profiles from different cell types, which may be correlated to a corresponding composition of the cell for identification purposes. The identification may be qualitative and / or quantitative.

[0045] Slow cell screening may be particularly disadvantageous. For instance, determining a cell type from a sample determined from a subject potentially having a disease may be a limiting step to determining an appropriate treatment therefor. In some such cases, the time to determine a cell type may be limited by the time it takes to isolate a cell of interest, and / or wait for the cell to grow (e.g., grow and replicate) to a degree suitable for performing the cell screening method. In view of this, certain embodiments as discussed herein are generally directed to the benefits of faster cell screening using the hyperspectral imaging or other techniques such as those described herein. In some embodiments, hyperspectral imaging may be relatively fast, when compared to typical techniques for cell identification. This may be the case, in some embodiments, because the methods acquiring hyperspectral profiles may necessitate fewer cells for identifying a cell type when compared to typical cell screening methods. For instance, in some embodiments, the systems and methods disclosed herein include acquiring a hyperspectral profile of a single cell. In some embodiments, the systems and methods disclosed herein include acquiring a hyperspectral profile of a single cell within a population of cells. Accordingly, the wait time for cell growth before identification may be reduced, in accordance with some embodiments, e.g., when compared to typical cell screening methods.

[0046] In some embodiments, the systems and methods described herein utilize microfluidic devices having one or more cell growth trenches. As described elsewhere herein, the cell growth trenches may facilitate separation and isolation of different cell types from a sample. In accordance with some embodiments, cells may be contained within the cell growth trenches, where hyperspectral or other profiles of one or more of the cells may then be acquired. In this manner, according to some embodiments, separation of cells, growth of the cells, and identification of the cells in a parallel manner may occur, which may desirably accelerate the process of identifying unknown cells from a sample, e.g., when compared to typical cell screening methods.

[0047] According to some embodiments, systems configured for hyperspectral imaging or other imaging techniques are described. Hyperspectral imaging of a cell generally refers to the process of collecting fluorescence emission spectra of the cell at one or more excitation wavelengths to obtain a hyperspectral profile (i.e., the collection of fluorescence emission spectral). The collection of the fluorescence emission spectra may be correlated to certain types and / or amounts of molecules present within the imaged cell, which may provide information regarding a composition of the cell and / or be correlated to a cell type. For example, in accordance with some embodiments, samples of cells (e.g., a single cell or a single type of cell) where an identity of the cell is known may be imaged to acquire corresponding hyperspectral profiles of the known cells. This may be repeated with various types of cells to form a library of hyperspectral profiles in certain embodiments. Such a library of hyperspectral profiles, in some embodiments, may be used to identify a cell type from a hyperspectral profile acquired from an unknown cell. As a non-limiting example, a hyperspectral profile may be collected of an unknown cell, and the hyperspectral profile may be correlated to a known cell type to identify the unknown cell. Other libraries can also be prepared in other embodiments, e.g., using other imaging techniques such as those described herein.

[0048] For example, while various embodiments disclosed herein are described in the context of hyperspectral imaging, it will be appreciated that other imaging techniques may be used in place of hyperspectral imaging. Non-limiting examples of other techniques include quantitative phase microscopy and fluorescence lifetime imaging, e.g., as discussed in patent applications filed on May 13, 2024, entitled “Microscopy Systems and Methods for Identifying Bacteria and Other Organisms in Microfluidic Devices” (U.S. Pat. Apl. Ser. No. 63 / 646,599) and “Identifying Bacteria and Other Organisms by Determining Growth Profiles” (U.S. Pat. Apl. Ser. No. 63 / 646,582). Each of these is incorporated herein by reference in its entirety.

[0049] Some aspects of the present disclosure are generally related to systems configured to perform hyperspectral imaging, or other imaging techniques. In some embodiments, the system may include a light source, a substrate configured to contain one or more cells, and a detector. FIG. 1A shows a non-limiting example of one such system configured for hyperspectral imaging, including a light source 102, a substrate configured to contain cells 108, and a detector 110.

[0050] In some embodiments, a light source may be positioned and oriented to direct light toward a substrate. In some cases, a particularly beneficial type of light source having a certain intensity and / or wavelength of light may be employed in the system. In one set of embodiments, the light source may have a visible wavelength spectrum. In one set of embodiments, the light source (e.g., UV laser) may have a wavelength in the ultraviolet regime. Other ranges of light may also be used in other embodiments. Non-limiting examples of light source that may be employed include, but are not limited to, a laser, a light-emitting diode, an arc lamp, etc. In some cases, more than one light source may be used, and these may be the same or different, e.g., providing the same or different wavelengths of light. In some embodiments, the light source is collimated, converging, and / or diverging.

[0051] In some embodiments, the light source (e.g., a laser) is configured to produce light having an intensity and / or wavelength capable of exciting one or more molecules present within one or more cells. For example, in some embodiments, the light source may be configured to produce light having an intensity of greater than or equal to 0.1 W / cm2, greater than or equal 0.5 W / cm2, greater than or equal 1 W / cm2, greater than or equal to 5 W / cm2, greater than or equal 10 W / cm2, greater than or equal to 50 W / cm2, greater than or equal 100 W / cm2, greater than or equal 500 W / cm2, greater than or equal to 1,000 W / cm2, greater than or equal 5,000 W / cm2, greater than or equal 10,000 W / cm2, or greater than or equal 50,000 W / cm2. In some embodiments, the light source may be configured to produce light having an intensity of less than or equal to 100,000 W / cm2, less than or equal 50,000 W / cm2, less than or equal 10,000 W / cm2, less than or equal 5,000 W / cm2, less than or equal 1,000 W / cm2, less than or equal to 500 W / cm2, less than or equal to 100 W / cm2, less than or equal to 50 W / cm2, less than or equal to 10 W / cm2, less than or equal to 5 W / cm2, less than or equal to 1 W / cm2, or less than or equal 0.5 W / cm2. Any of the above reference ranges are possible (e.g., greater than or equal to 0.1 W / cm2and less than or equal to 100,000 W / cm2, or greater than or equal to 100 W / cm2and less than or equal to 1,000 W / cm2). Other ranges are also possible.

[0052] In some embodiments, the light source (e.g., a laser) is configured to a produce light having any of a variety of appropriate wavelengths. In some cases, the light source is able to produce continuous spectral bands, e.g., for use in various hyperspectral imaging techniques. In some embodiments, the system may include one or multiple light sources, e.g., to excite cells with different wavelengths. In some embodiments, a light source (e.g., a laser) may have a wavelength of greater than or equal to 100 nm, greater than or equal to 125 nm, greater than or equal to 150 nm, greater than or equal to 180 nm, greater than or equal to 200 nm, greater than or equal to 225 nm, greater than or equal to 250 nm, greater than or equal to 275 nm, greater than or equal to 300 nm, greater than or equal to 325 nm, greater than or equal to 350 nm, greater than or equal to 375 nm, greater than or equal to 400 nm, greater than or equal to 420 nm, greater than or equal to 440 nm, greater than or equal to 450 nm, greater than or equal to 480 nm, greater than or equal to 500 nm, greater than or equal to 550 nm, greater than or equal to 600 nm, greater than or equal to 650 nm, greater than or equal to 700 nm, greater than or equal to 800 nm, or greater than or equal to 900 nm. In some embodiments, the light source (e.g., a laser) may have a wavelength of less than or equal to 1000 nm, less than or equal to 900 nm, less than or equal to 800 nm, less than or equal to 700 nm, less than or equal to 650 nm, less than or equal to 600 nm, less than or equal to 550 nm, less than or equal to 500 nm, less than or equal to 480 nm, less than or equal to 450 nm, less than or equal to 440 nm, less than or equal to 420 nm, less than or equal to 400 nm, less than or equal to 375 nm, less than or equal to 350 nm, less than or equal to 325 nm, less than or equal to 300 nm, less than or equal to 275 nm, less than or equal to 250 nm, less than or equal to 225 nm, less than or equal to 200 nm, or less than or equal to 180 nm, less than or equal to 150 nm, or less than or equal to 125 nm. The above-referenced values of wavelengths may have a deviation of + / - 5 nm, of + / - 10 nm, or + / - 15 nm. Any of the abovereferenced ranges may be possible (e.g., greater than or equal to 100 nm + / - 5 nm and less than or equal to 1,000 nm + / - 5 nm, greater than or equal to 200 nm + / - 5 nm and less than or equal to 480 nm + / - 5 nm, or greater than or equal to 250 nm + / - 5 nm and less than or equal to 400 nm + / - 5 nm). Other ranges are also possible.

[0053] In some embodiments, the system may be configured to focus light to or near the diffraction limit. Accordingly, in some embodiments, the light source may be coherent as noted above (e.g., a laser). In some embodiments, the coherent light source has a high quality, e.g., with an M2value of less than or equal to 1.2, less than or equal to 1.18, less than or equal to 1.16, less than or equal to 1.14, less than or equal to 1.12, less than or equal to 1.1, less than or equal to 1.08, or less.

[0054] In some embodiments, the system may include a structuring element. Non-limiting examples of structuring elements include a microlens array, a cylindrical lens, a Powell lens, a prism, a mirror, a beamsplitter, a diffraction grating, or other suitable lenses or other elements. In some embodiments, the light source may not be coherent, and in some cases, a mask and / or a focusing element may be used, for example, to focus the light, e.g., to or near the diffraction limit. In accordance with some embodiments, the light source comprises a structured light source such that light from the light source is directed to only a portion of a sample (e.g., a portion of microfluidic device). In some embodiments, a structured light source illuminates only a portion of the microfluidic device, e.g., one or more cell growth trenches as described elsewhere herein in more detail. In some embodiments, the structured light source may be scanned to illuminate (e.g., an accordingly image) different portions of the microfluidic device. For instance, in some embodiments, the structured light source is configured with one or more components such that it is capable of separately illuminating and facilitating imaging a first portion and then a second portion of a sample (e.g., a microfluidic device). In some such embodiments, non-limiting components that may facilitate scanning different portions of a sample include a galvanometer scanner and acousto-optical devices. In addition, in some embodiments, it may be possible to move the sample while keeping the illumination fixed. Combinations of these approaches are also possible in some cases. Other such components arc possible, and may be selected based on the insight provided by this disclosure.

[0055] In some embodiments, focused light may facilitate hyperspectral imaging, e.g., of a single cell. In accordance with some embodiments, the light source is focused to or near the diffraction limit when exposed to cells. In some embodiments, the light source is focused to or near the diffraction limit at a single point (e.g., to excite a single cell). In some embodiments the light source is focused to or near the diffraction limit to an array of spots, a line, an array of lines, or any other suitable 0-D, 1-D, or 2-D patterns (e.g., orthogonal lines). The focused light in some cases may facilitate sufficient emission from one or more excited cells, for instance, to be dispersed by a dispersing element or a structuring element so that at least some, and in some cases all, of the constituent wavelengths of the emission may be measured at a detector, in accordance with some embodiments. In some embodiments, for example, a prism may be the dispersing element. In some cases, the prism may be positioned and oriented to avoid overlapping spectra. In some embodiments, the prism may be placed in the Fourier space or image space in the emission light path.

[0056] Any of a variety of suitable substrates are possible in the systems described herein, in some embodiments. In some embodiments, the substrate may be transparent to wavelengths of light that may be used to excite cells and / or emitted from cells (e.g., as auto fluorescence). For instance, in some cases, the substrate may comprise glass, optically transparent polymers (e.g., PDMS), or the like. Non-limiting examples of suitable substrates include a microscope slide and coverslip, a petri dish, and / or a microfluidic device. In accordance with some embodiments, and as described elsewhere herein, the substrate configured to contain cells may be a microfluidic device including one or more cell growth trenches. Other substrates are also possible. In some embodiments, the substrate configured to contain cells includes the cells. In some such embodiments, at least some of the cells are an unknown type of cell.

[0057] A detector, in some embodiments, is included in the system to acquire the hyperspectral profile from the cells. Any of a variety of suitable detectors are possible, including charge coupled devices (CCDs), complementary metal-oxide semiconductor (CMOS) sensors, photocathodes, photomultiplier tubes, or the like. In some embodiments, the detector comprises a color sensitive detector. In some cases, the detector comprises a non-color sensitive detector configured with one or more wavelength sensitive optical elements, such as dispersive elements (e.g., prisms, diffraction gratings, or the like) and / or filters (e.g., emission filters or acousto- optical devices).

[0058] Processors, in accordance with some embodiments, may be included within the system. In some embodiments, a processor may be configured and arranged to perform hyperspectral analysis on the fluorescence received by the detector.

[0059] Other typical optical elements may be included, in some embodiments, to facilitate hyperspectral imaging. In some embodiments, mirrors, beam splitters, optical filters, prisms, etc. may be included in the system. In some embodiments, additional optical elements may facilitate hyperspectral imaging. For example, a first excitation wavelength may be directed toward a cell, which then produces an emissive spectrum (e.g., via autofluorescence). The emissive spectrum may then be measure at a variety of distinct wavelengths using optical filters (e.g., a filter cube) in a sequential manner. In some embodiments, the emissive spectrum may be dispersed by a dispersing element or a structuring element, e.g., as described above, and may be measured in some cases in a continuous manner (e.g., all the wavelengths detected with a detector substantially simultaneously). Discrete or non-continuous measurements are also contemplated in other embodiments. Measuring the emissive spectra of a cell (e.g., sequentially or continuously, etc.) may be repeated using a second excitation wavelength, and optionally, a third excitation wavelength, a fourth excitation wavelength, and so forth, in accordance with some embodiments.

[0060] In some embodiments, the system may comprise a mask (e.g., an electronically reconfigurable mask) positioned to selectively shield the light directed (by a light source) at at least a portion of the substrate. In some embodiments, where the substrate comprises a microfluidic device, the mask may be positioned to selectively shield the light directed at at least a portion of the microfluidic device. The mask may be manually or automatically controlled.

[0061] The system described herein may comprise any of a variety of appropriate types of masks. Non-limiting examples of masks include a spatial light modulator, a fixed aperture, a filter, etc. In some embodiments, the spatial light modulator comprises a micro-mechanical mirror-based spatial light modulator. In some cases, the spatial light modulator may include, for example, a digital micromirror device (DMD), a ferroelectric liquid crystal on silicon (LCOS) chip, a nematic liquid crystal (NLC) platform, a grating light valve (GLV), etc. In addition, in certain embodiments, more than one mask may be present, and the masks may independently be the same or different. In certain embodiments, devices comprising a spatial light modulator, for example, a digital micromirror device, may facilitate the application of hyperspectral imaging to a single cell. In certain embodiments, devices comprising a spatial light modulator facilitate the application of hyperspectral imaging to a single cell within a population of cells. Most hyperspectral light sources illuminate areas that are substantially larger than single cells, and accordingly hyperspectral imaging techniques have not previously been able to image or analyze a single cell within a population of cells.

[0062] According to some embodiments, as mentioned above, the systems described herein may include a microfluidic device. In some embodiments, the microfluidic device may be configured to contain cells. In some embodiments, a system configured for individual cell screening is disclosed herein. In one set of embodiments, the system comprises a microfluidic device configured to contain a plurality of cells and a source of light (e.g., a laser) configured to induce autofluorescence in at least a portion of the plurality of cells (e.g., excite at least some of the molecules within at least a portion of the plurality of cells). A non-limiting example of the systems described herein is shown in FIG. IB. As shown, system 1 comprises the microfluidic device 4, light source 2 (e.g., a laser) positioned to direct light at at least a portion of the microfluidic device, and mask 3 (e.g., an electronically reconfigurable mask) positioned to selectively shield a least a portion of the light directed at the portion of the microfluidic device 4. The mask, in some cases, may be configured to shield a portion 7 of the light originating from the light source 2, such that the portion of the microfluidic device underneath the mask is prevented from being exposed to light. As a result, cells residing within the portion of the microfluidic device may be shielded from light.

[0063] In some embodiments, the microfluidic device comprises two or more layers comprising microfluidic channels. For example, the microfluidic device may comprise a cell flow layer configured for flowing, receiving, and housing a plurality of cells. In some cases, the microfluidic device may further comprise a control layer positioned adjacent (e.g., coupled to) the cell flow layer. The control layer may be configured to control the flow of a fluid (e.g., cell media, a fluid comprising the plurality of cells, etc.) in the cell flow layer. In some embodiments, the microfluidic device further comprises a substrate (e.g., a coverslip) coupled to the cell flow layer, thereby forming a base layer of the microfluidic device. As shown in the non-limiting example of the microfluidic device in FIG. IB, the microfluidic device 4 may comprise a cell flow layer 10 and a control layer 50 positioned adjacent the cell flow layer 10. A substrate 8 may be coupled to the cell flow layer 10, forming the base layer of the microfluidic device 4.

[0064] While FIG. IB shows one embodiment in which the microfluidic device comprises three layers, it should be noted that not all embodiments described herein are so limiting, and in other embodiments, the microfluidic device may comprise any additional layers disposed adjacent (e.g., directly adjacent) and / or between the one of more layers described above. In other embodiments, more or fewer layers may be present.

[0065] As noted above, the microfluidic device may include a substrate (e.g., coverslip), a cell flow layer coupled to the substrate, and a control layer coupled to the cell flow layer, in one embodiment. In some embodiments, the cell flow layer may include a variety of different channels through which cells (and other fluids such as growth media and cleaning fluids or solutions) can flow during use. The control layer may include various channels that can be filled with a fluid in order to actuate various different portions (e.g., valve portions) of the cell flow layer. In some embodiments, the cell flow layer may comprise various portions (e.g., valve portions) that can be actuated by the control layer to selectively control the flow of cells and other fluids through the various channels of the cell flow layer. It should be understood, however, that flow layers, control layers, etc. in this example are but one method of confining or growing cells on a microfluidic device, e.g., for screening or other purposes, but that in other embodiments, other systems for confining or growing cells in a microfluidic device are also contemplated.

[0066] In some embodiments, the microfluidic device may comprise various channels having any of a variety of configurations and arrangements described herein. A non-limiting example of one embodiment is shown in FIGs. 2A-2C. As shown, FIGs. 2A-2C respectively illustrate a top down view, perspective view, and a partially exploded perspective view of the microfluidic device 4 of FIG. IB.

[0067] In one set of embodiments, the microfluidic device includes a cell flow layer comprising one or more growth channels configured to receive, flow, and / or house a plurality of cells. As shown in FIGs. 2A-2C, the microfluid device 4 may comprise a cell flow layer 10 comprising one or more growth channels 12. In some embodiments, the control flow layer comprises one or more control channels configured to control flow of a fluid in the cell flow layer. For example, as shown in FIGs. 2A-2C, the control flow layer 50 may comprise one or more control channels 52A configured to control flow of a fluid in the one of more growth channels 12 within the cell flow layer 10.

[0068] While FIGs. 2A-2C show a set of embodiments in which the microfluidic device comprises growth channels and control channels, it should be noted that not all embodiments described herein are so limiting, and in other embodiments, the microfluidic device may comprise various other types of channels in the cell flow layer and / or control layer. For example, as described in more detail below, the cell flow layer may further comprise one or more collection channels in additional to the growth channels.

[0069] In some embodiments, the channels (e.g., growth channels) in the cell flow layer are separated from the channels (e.g., control channels) in the control layer by an upper wall of the cell flow layer. See also Int. Pat. Apl. Pub. No. WO 2020 / 257746. As shown in FIG. 2C, the various channels (e.g., growth channels 12) of the cell flow layer 10 may be defined on an underside of the cell flow layer 10. For instance, the cell flow layer 10 may include an upper wall 10A that forms the upper wall (e.g., the ceiling) of the various channels defined in the cell flow layer 10. When the cell flow layer 10 is bonded to the substrate 8, the substrate 8 forms a lower wall (e.g., a floor) of the various channels of the cell flow layer 10. Similarly, the various channels (e.g., control channels) of the control layer 50 may be defined on an underside of the control layer 50. An upper wall 50A of the control layer 50 may form an upper wall (e.g., a ceiling) of the various channels (e.g., control channels) of the control layer 50, and the upper wall 10A of the cell flow layer 10 may form a lower wall (e.g., a floor) of the various channels of the control layer 50.

[0070] In some embodiments, the channels of the cell flow layer and the control layer may be fhiidically coupled to the atmosphere via a plurality of vertical channels or ports. For example, as shown in FIGs. 2A-2C, the various channels (e.g., growth channel 12) of the cell flow layer 10 may be fluidically coupled to the atmosphere via a plurality of vertical channels extending upward through the cell flow layer 10 and the control layer 50 via one or more openings 15 and 17 defined in the upper wall 50A of the control layer 50. Similarly, the various channels (e.g., control channels 52A) of the control layer 50 may be fluidically coupled to the atmosphere via a plurality of vertical channels extending upward through the control layer 50 via one or more openings 19 defined in the upper wall 50A of the control layer 50. In some embodiments, the one or more openings may serve as inlets and / or outlet that allow for fluid flow into or out of the various channels in the control layer and cell flow layer.

[0071] In some embodiments, the cell flow layer comprises a growth channel comprising various portions, including an inlet portion, an outlet portion, and a main portion positioned between the inlet portion and the outlet portion. The growth channel may further comprise an inlet valve portion associated with the inlet portion and an outlet valve portion associated with the outlet portion. The various portions of the growth channel may be located at various positions along the length of the cell flow layer. For example, the inlet portion and the inlet valve portion may be located at a first end of the cell flow layer, while the outlet valve portion and the outlet portion may be located at a second end of the cell flow layer. The main portion may be located between the first end and the second end of the cell flow layer. Generally, each portion of the growth channel may function as a channel through which cells and fluids can flow.

[0072] A non-limiting example of one embodiment of a growth channel is illustrated in FIG. 3A. Specifically, FIG. 3A shows a top-down view of a section (e.g., section 4A) of the same microfluidic device illustrated in FIG. 2A. As shown, the growth channel 10 comprises various portions, including an inlet portion 14A at a first end of the cell flow layer, an outlet portion 14B at a second end of the cell flow layer, and a main portion 18 positioned between the inlet portion 14A and the outlet portion 14B. The growth channel 10 may further comprise an inlet valve portion 16A associated with the inlet portion 14A and an outlet valve portion 16B associated with the outlet portion 14B.

[0073] In some embodiments, the cell flow layer further comprises a plurality of cell growth trenches fluidically coupled to a main portion of the growth channel. In some embodiments, the plurality of cell growth trenches is configured to contain cells during use of the device. As shown in FIG. 3A, the cell flow layer may comprise a plurality of cell growth trenches 20 fluidically coupled to the main portion 18 of the growth channel 10. In some cases, the cell growth trenches may be positioned on a first side of the main portion of the growth channel, and may be configured to extend outward from the main portion of the growth channel in a direction that is perpendicular to the direction in which the main portion extends between a first end of and a second end of the cell flow layer. For example, as shown in FIG. 3A, the plurality of cells growth trenches 20 may be positioned on a first side of the main portion 18 of the growth channel and may extend outward from the main portion 18 in a direction perpendicular to a direction in which the main portion 18 extends. However, it should be understood that other arrangements of cell growth trenches are also possible in other embodiments. For example, the cell growth trenches may appear on both sides of growth channel, there may be more than one such growth channel, the cell growth trenches may independently be of the same or different sizes or lengths, the cell growth trenches may be symmetrically or asymmetrically arranged within the cell flow layer, or the like.

[0074] In some embodiments, the microfluidic device comprises a plurality of single-entry, single-file cell growth trenches fluidically coupled to the main portion of the growth channel. That is, some or all of the plurality of cell growth trenches may be sized so that individual cells are permitted to enter into the cell trenches one at a time in a linear-, single-file fashion. In some cases, the cell growth trenches (e.g., growth trenches 20 in FIG. 3A) may have a width (e.g., width w in FIG. 3A) that is generally equal to or slightly larger than the width of individual cells entering into the growth trenches. For example, as shown in FIG. 4B, the single-entry, singlefile cell growth trenches 20 may be sized such that individual cells are configured to fill the trenches in a linear-, single-file fashion. In addition, in other embodiments, some or all of the plurality of cell growth trenches may be sized to allow more than one individual cell to enter at a time.

[0075] In some cases, as noted above, the system may include a mask, which may be positioned to selectively shield light directed at at least a portion of the cell growth trenches in the cell flow layer of a microfluidic device. For example, light may be directed at the cells that may be sufficient to inhibit or kill the cells. However, the light may not necessarily reach all of the cells in the device due to the presence of the mask, which may be able to selectively shield the incoming light. By positioning or configuring the mask appropriately, in some embodiments, certain cells or portions of the device may be subjected to light, while other cells or portions of the device may be shielded from the light, e.g., such that those cells survive. In some embodiments, this may be advantageous as the light may inadvertently kill cells exposed thereto, but the cell growth within the cell growth trenches may facilitate the “pushing” out of dead cells, as described elsewhere herein. It should additionally be understood that, in some embodiments, the light incident on the cells may not kill the cells. Accordingly, in some such embodiments, the cell screening method may be non-destructive, and cells may then be separated and / or collected following identification.

[0076] As mentioned above, the system may include a mask positioned to shield at least a portion of the substrate from the light source. In some embodiments, the mask may be configured to allow light to be incident on only a portion of the substrate. For example, in some embodiments, a mask may be configured to allow light to be incident on a microfluidic device in a line that is perpendicular to a direction of one or more cell growth trenches of the microfluidic device, e.g., as shown as dotted line 51 in FIG. 3B. In this manner, some or all of the cells from each of the cell growth trenches containing cells may be excited by the light source, such that they may autofluoresce and hyperspectral profiles of cells from each populated cell growth trench may be collected and analyzed in parallel by the detector and / or processor, in accordance with some embodiments. Such an arrangement, in some embodiments, may desirably facilitate high throughput screening of cells within a microfluidic device using hyperspectral imaging.

[0077] As noted above, the growth channel (e.g., 10 in FIG. 3A) may include, in certain embodiments, various valve portions, e.g., such as an inlet valve portion (e.g., 16A in FIG. 3A) and an outlet valve portion (e.g., 16B in FIG. 3A). In some embodiments, these valve portions may be actuated to aid in selectively controlling the flow of cells and fluid through the growth channel in the cell flow layer. For instance, the inlet valve portion of the growth channel may allow for control of fluid flow between the inlet portion of the growth channel and the main portion of the growth channel. Similarly, the outlet valve portion of the growth channel may allow for control of fluid flow between the main portion of the growth channel and the outlet portion of the growth channel. For example, as shown in FIG. 3 A, the inlet valve portion 16A of the growth channel 10 may allow for control of a flow 22 A between the inlet portion 14A of the growth channel 10 and the main portion 18 of the growth channel 10. Similarly, the outlet valve portion 16B of the growth channel 10 may allow for control of a flow 22B between the main portion 18 of the growth channel 10 and the outlet portion 14B of the growth channel 10. See also Int. Pat. Apl. Pub. No. WO 2020 / 257746. However, it should also be noted that in other embodiments, there may be only a single valve portion present (e.g., an inlet valve portion or an outlet valve portion), and / or there may be no valve portions present.

[0078] In some embodiments, in the microfluidic device, the control layer may include one or more control channels configured to actuate the valve portions (e.g., inlet and / or outlet valve portions) within the growth channel in the cell flow layer. A variety of microfluidic valve configurations may be used in various embodiments.

[0079] A non-limiting example of one such embodiment is illustrated in FIG. 3A. As shown, the control layer (e.g., control layer 50 in FIG. 2B) may include one or more control channels 52A coupled to (e.g., overlaps with) the inlet valve portion 16A and outlet valve portion 16B of the growth channels 10 in the cell flow layer (e.g., cell flow layer 10 in FIG.2). The control channels 52A may be configured to aid in actuating the inlet valve portion 16A of the growth channel 10 and the outlet valve portion 16B of the growth channel 10. The control channels may have any of a variety of appropriate configurations in the control layer. For example, in one set of embodiments, the control channel may have a U shape, as shown in FIGs. 2A-2C. As shown, the control layer may include one or more control channels 52A fluidically connected by a base channel 52B, thereby forming an overall U-shaped structure. In this embodiment, the control channels 52A may extend across (e.g., overlap with) the inlet valve portion 16A and outlet valve portion 16B of the growth channels 10. Specifically, the control channel 52A may overlap with portions of the upper wall 10A of the cell flow layer 10 that form the upper wall of the inlet valve portions 16A and outlet valve portion 16B.

[0080] While FIGs. 2A-2C illustrate an embodiment in which the control channels have an overall U shape structure, it should be noted that the control channels can have other shapes and / or configurations. In some embodiments, the control channels may extend across (e.g., overlaps with) all of the necessary valve portions of the cell flow layer. By having such a configuration, the control channel may serve as an on-off switch configured to close or open the valve portions of the cell flow layer. In addition, it should be understood that other valve configurations are also possible in other embodiments, for example, valve configurations that are able to control flow proportionally.

[0081] In some embodiments, one or more control channels in the control layer can be pressurized in order to actuate the valve portions (e.g., inlet and / or outlet valve portions) of the growth channels in the cell flow layer, for example, to serve as an on-off switch, or such that flow through the valve is proportional to the amount of pressure. As one non-limiting example, when the one or more control channels become pressurized, the inlet and outlet valve portions of the growth channels may be actuated to transition from an open state that allows for fluid flow through the valve portions to a closed state that prevents fluid from flowing through the valve portions. Conversely, when the one or more control channels become depressurized, the inlet and outlet valve portions of the growth channels may transition back from a closed state (i.e., actuated state) that prevents fluid from flowing through the valve portions to an open state (i.e., unactuated state) that allows for fluid flow through the valve portions.

[0082] FIGs. 3A-3B shows a non-limiting example of actuation of the inlet and outlet valve portions via the one or more control channels. As shown in FIG. 3A, when the one or more control channels 52A are not pressurized, the inlet and outlet valve portions 16A and 16B of the growth channels 12 remain in an open state that allows for fluids 22A and 22B to flow through the valve portions 16A and 16B. As shown in FIG. 3B, when the one or more control channels 52A become pressurized, the inlet and outlet valve portions 16A and 16B of the growth channels 12 become actuated and transition to a closed state that prevents fluid from flowing through the valve portions. Conversely, when the one or more channels become depressurized, the inlet and outlet valve portions 52A of the growth channels may revert from a closed state (as shown in FIG. 3B) to an open state (as shown in FIG. 3A).

[0083] In some embodiments, the cell flow layer and / or the control layer comprise polymers (e.g., polydimethylsolixane (PDMS)) and may be cast together, or from separate molds. In some embodiments, the substrate is made from glass. The various channels of the cell How layer and the control layer can be formed using any suitable fabrication technique(s). In some embodiments, the cell flow layer and the control layer are fabricated using multilayer soft lithography. In some embodiments, molds are initially formed from silicon wafers using UV lithography techniques. The cell flow and / or control polymer layers may then cast by flowing liquid polymer into the silicon molds, and then subsequently cured so that the polymer hardens. The two polymer layers can be bonded together (for example via curing or partial curing), and bonded to the substrate (for example via plasma bonding), and then further baked. Thus, the negative space of the channels of the cell flow layer and the control layer may be imprinted from the positive silicon wafer molds.

[0084] In some embodiments, the cell flow layer may have a length (e.g., such as a distance between the inlet and outlet of each channel in the cell flow layer) of between about 5 mm and about 100 mm, or about 30 mm; the control layer may have a span between various control channels (e.g., such as a distance between the two control channels 52A in FIG. 2A) of between about 4 mm and about 99 mm, or about 29 mm. The cell flow layer 10 and the control layer 50 may independently have a width of between about 20 micrometers and about 500 micrometers, or about 100 micrometers; the cell flow layer 10 may have a height of between about 5 micrometers and about 80 micrometers, or about 15 micrometers; and the control layer 50 may have a height of between about 10 micrometers and about 100 micrometers, or about 50 micrometers. The cell growth trenches may have any of a variety of appropriate dimensions. In some embodiments, the length (e.g., 1 in FIG. 3A) of the cell growth trenches (e.g., the distance that the cell growth trenches 20 extend outward from the main portion 18 of growth channel 10) may be at least 1 micrometer, at least 5 micrometers, at least 10 micrometers, at least 25 micrometers, at least 50 micrometers, at least 75 micrometers, at least 100 micrometers, at least 200 micrometers, at least 300 micrometers, or at least 400 micrometers. In some embodiments, the length of the cell growth trenches may be no more than 500 micrometers, no more than 400 micrometers, no more than 300 micrometers, no more than 200 micrometers, no more than 100 micrometers, no more than 75 micrometers, no more than 50 micrometers, no more than 25 micrometers, no more than 10 micrometers, or no more than 5 micrometers. Any of the abovereferenced ranged are possible (e.g., between 1.0 micrometer and 500.0 micrometers). Other ranges are also possible.

[0085] As noted above, the width (e.g., w in FIG. 3A) of the cell growth trenches (e.g., 20 in FIG. 3A) may be comparable to the size of a single cell in certain embodiments. For example, the cell growth trench may have a width that is large enough fit a single cell. In some embodiments, the width of the cell growth trench may be at least 1 micrometer, at least 5 micrometers, at least 10 micrometers, at least 25 micrometers, at least 50 micrometers, or at least 75 micrometers. In some embodiments, the length of the cell growth trenches may be no more than 100 micrometers, no more than 75 micrometers, no more than 50 micrometers, no more than 25 micrometers, no more than 10 micrometers, or no more than 5 micrometers. Any of the above-referenced ranged are possible (e.g., between 1.0 micrometer and 100.0 micrometers). Other ranges are also possible.

[0086] In some embodiment, the height of the cell growth trenches may be at least 0.1 micrometers, at least 0.5 micrometers, at least 1 micrometer, at least 5 micrometers, at least 10 micrometers, at least 20 micrometers, or at least 40 micrometers. In some embodiments, the length of the cell growth trenches may be no more than 50 micrometers, no more than 40 micrometers, no more than 20 micrometers, no more than 10 micrometers, no more than 5 micrometers, no more than 1 micrometers, or no more than 0.5 micrometers. Any of the abovereferenced ranged are possible (e.g., between 0.1 micrometer and 50.0 micrometers). Other ranges are also possible. In some embodiment, the distance between adjacent pair of cell growth trenches (e.g., 20 in FIG. 3A) may be at least 0.1 micrometers, at least 0.5 micrometers, at least 1 micrometer, at least 2 micrometers, at least 4 micrometers, at least 6 micrometers, at least 8 micrometers, at least 10 micrometers, at least 20 micrometers, at least 30 micrometers, or at least 40 micrometers. In some embodiment, the distance between adjacent pair of cell growth trenches may be no more than 50 micrometers, no more than 40 micrometers, no more than 30 micrometers, no more than 20 micrometers, no more than 10 micrometers, no more 8 micrometers, no more 6 micrometers, no more 4 micrometers, no more 2 micrometers, no more 1 micrometers, or no more 0.5 micrometers. Any of the above-referenced ranged are possible (e.g., between 0.1 micrometer and 50.0 micrometers). Other ranges are also possible.

[0087] The cell growth trenches may have any of a variety of appropriate aspect ratios. For example, in one set of embodiments, the cell growth trenches may have a length (e.g., 1 in FIG. 3A) to width (e.g., w in FIG. 3A) aspect ratio of at least 2, at least 5, at least 10, at least 25, at least 50, or at least 75. In some embodiments, the cell growth trenches may have a length to width aspect ratio of no more than 100, no more than 75, no more than 50, no more than 25, no more than 10, or no more than 5. Any of the above-referenced ranged are possible (e.g., between 2 and 100). Other ranges are also possible.

[0088] Some aspects are related to methods of using the systems described herein. In some cases, the methods may include providing a cell. In some embodiments, the cell is a known or an unknown species. In some embodiments, the method includes providing a plurality of cells. In some embodiments, at least some of the cells of the plurality of cells are of an unknown species. In accordance with some embodiments, the cells comprise one or more different unknown species of unicellular organisms. In some embodiments, providing the cells may comprise acquiring the cells, e.g., by sampling cells from a source. According to some embodiments, the sample may be obtained from any of a variety of suitable sources, e.g., from which it may be desirable to know the identity of the unknown cells. For instance, in some embodiments, the sample may be obtained from the environment (e.g., a potentially contaminated body of water, soil, etc.), from a food source, and / or a subject having a disease. The subject may be human or a non-human animal, e.g., a non-human mammal. The unknown cells may be any of a variety of suitable cells suitable for identification by hyperspectral imaging. Non-limiting examples of cells include bacteria cells, mammalian cells, fungi, algae, protozoa, archaea, etc.

[0089] In some embodiments, at least some of the cells of the plurality of cells fluoresce (e.g., auto fluoresce). In some embodiments, the autofluorescence is detected by the systems described herein. For instance, in some embodiments, the method includes detecting at least of portion of light emitted as autofluorescence from at least some of the plurality of cells. In some embodiments, the autofluorescence of cells within the plurality of cells is weak, and performing hyperspectral imaging thereof can be challenging. Advantageously, in some embodiments, the methods include providing one or more compounds to the plurality of cells to enhance autofluorescence. In some embodiments, the one or more compound comprises 5-aminolevulinic acid (5-ALA) and / or shikimate. Other compounds are possible, and may be selected based on the identity or class (e.g., bacterial, fungal, algal, etc.) of the cells within the plurality of cells.

[0090] In some embodiments, the method comprises providing a substrate, as described elsewhere herein. In some embodiments, the substrate comprises a microfluidic device. As a non-limiting example, as noted above with respect to FIGs. 1B-2C, the microfluidic device may comprise, in one embodiment, a substrate 8, a cell flow layer 10 coupled to the substrate 8, and a control layer 50 configured to control flow of the fluid in the cell flow layer 10. In certain embodiments, the cell flow layer may comprise a growth channel having an inlet portion, an outlet portion, an inlet valve portion associated with the inlet portion, a main portion, and an outlet valve portion associated with the outlet portion. The microfluidic device and associated components (e.g., cell flow layer, control layer, growth channels, control channels, cell growth trenches, etc.) may have any configurations and properties described previously. Non-limiting examples include those described with reference to FIGs. 1-3.

[0091] In some embodiments, the method comprises adding or injecting a fluid comprising a plurality of cells into the microfluidic device. Accordingly, in some embodiments, the method includes introducing a plurality of cells into a microfluidic device, e.g., that comprises one or more cell growth trenches. In some embodiments, the cells and the growth media may be flowed into a main portion of the growth channel. For example, as shown in FIG. 4A, cells 74 and growth media may be injected into the growth channel 12 through the inlet portion 14A. In some cases, the cells and growth media can be injected via one or more of the inlet openings. For example, as shown in FIG. 4A, cells 74 and growth media can be injected via one or more inlet openings 15 A and 15B and flow into the main portion 18 of the growth channel 12. Any of a variety of cells may be injected into the microfluidic device. Non-limiting examples of cells include bacteria cells, mammalian cells, fungi, algae, protozoa, archaea, etc.

[0092] In some embodiments, during the injection step, the inlet and / or outlet valve portions of the growth channels are in an open state (e.g., an unactuated state). As noted above, in some embodiments, the control channels in the control layer are unpressurized. The control channels, when unpressurized, may allow fluid to flow through the valve portions of the growth channels. For example, as shown in FIG. 4A, when the control channel 52A is unpressurized, the inlet valve portions 16A and 16B are in an open state. The cells and growth media may thus able to flow through the inlet valve portion 16A into the main portion 18 of the growth channel 12.

[0093] In some embodiments, the cells injected into the main portion of growth channel may fill at least one of the plurality of cell growth trenches fluidically coupled to the growth channel. As shown in FIG. 4A, for example, the cells 74 injected into the main portion 18 may fill into at least one of the plurality of growth trenches 20. Any of a variety of methods may be employed to populate the cells from the main portion of the growth channel into the growth trenches. In some cases, for example, the cells may populate the cell growth trenches via diffusion. Additionally or alternatively, centrifugation may be employed to load the cell growth trenches.

[0094] In some embodiments, the method comprises providing cells contained with a plurality of single-entry, single-file cell growth trenches fluidically coupled to a growth channel. For example, after the injecting step, the plurality of single-entry and single-file cell growth trenches in the growth channel may be configured to contain the injected cells arranged in a linear, onedimensional grouping (e.g., the cells are geometrically constrained to a single-file line). See, for example, growth trenches in FIG. 4B. In some cases, as cells that initially fill the cell growth trenches begin to divide, the cells grow into an isogenic lineage of cells within the plurality of cell growth trenches. These cells within a given trench may have the same genetic makeup, e.g., as they may have originated from the same original cell. In some embodiments, the cells may be arranged in a single one-dimensional line within a trench. For example, as shown in FIG. 4B, as the cells 74 in each of the cell growth trenches 20 begin to divide, each of the cell growth trenches 20 may eventually contain an isogenic lineage of cells. In some embodiments, if there is more than one cell type present within a cell growth trench (e.g., a plurality of cells having different genetic lineages), then the cells may be allowed to expand and divide. The cell closest to the closed end of the channel may divide to produce additional cells, which “push” the other cells out of the channel (e.g., the ones having different genetic lineages) such that eventually, only an isogenic lineage of cells is present within the cell growth trench, originating from the cell closest to the closed end of the channel. In such a manner, according to some embodiments, each of a plurality of cell growth trenches initially containing cells may, after cell growth (e.g., replication, growth and division), contain and / or be filled by a single type of cell, e.g., after a cell initially closest to the closed end of the cell growth trench replicates and “pushes” any other cells out of the trench that were initial present. In this manner, according to some cases, cells injected into the microfluidic device may be separated and isolated (e.g., a cell type may be isolated after “pushing” out other cell types) within the cell growth trenches of the microfluidic devices.

[0095] In some embodiments, the method further comprises flushing the microfluidics device with a fluid, e.g., to remove the majority of cells from the growth channel while retaining one or more cells within at least one of the plurality of cell growth trenches. Referring again to FIG. 4B, as a non-limiting example, after the cells 74 populates the one or more of the cell growth trenches 20, a fluid may be injected through the main portion 18 of the growth channel 12 to flush out the majority of cells from the growth channels 12 while retaining the one or more cells 74 in the cell growth trenches 20.

[0096] Any appropriate fluids may be employed to flush one or more of the growth channels within the device, in various embodiments. Non-limiting examples of suitable fluids include cell media, buffer solutions, etc.

[0097] According to some embodiments, the method may include acquiring a hyperspectral profile or other imaging profile of a single cell. For example, the systems described herein, in some embodiments, may be designed and configured to facilitate hyperspectral imaging a single cell. In some embodiments, the method may include acquiring hyperspectral profiles of at least some of a plurality of cells. In some embodiments, the plurality of cells may be present in a microfluidic device. In some embodiments, the sample of cells may be provided on any of a variety of suitable substrates, as described elsewhere herein. In some such embodiments, the method may include acquiring a hyperspectral profile of the cells, at least one of which is of an unknown species. In some cases, acquiring a hyperspectral profile may include exposing the cells to a source of light having a first excitation wavelength and measuring an emission at one or more wavelengths.

[0098] In some embodiments, measuring an emission at one or more wavelengths from the exposed cells may be performed sequentially using a number of optical filters to measure the emissive light at certain wavelengths. In some such embodiments, measuring an emission at one or more wavelengths from the exposed cells may be performed continuously by dispersing the light with a dispersing element and measuring the dispersed light at a detector. For instance, the fluorescence emitted from the exposed cells, in some embodiments, may be refracted by a prism, whereafter the constituent emissive wavelengths may be simultaneously measured using a detector.

[0099] In some embodiments, acquiring a hyperspectral profile, or other imaging profile, may include exposing the cells to a light source having a second excitation wavelength (e.g., a different light source, the same light source with a different wavelength selected, etc.) and measuring a second emission at one or more wavelengths. This process, in some embodiments, may continue for a third excitation wavelength, a fourth excitation wavelength, and so forth to collect the hyperspectral profile. In some embodiments, as described elsewhere herein, at least a portion of the substrate containing the cells (e.g., microfluidic device) may be shielded by a reconfigurable mask. Accordingly, in some embodiments, the method may include shielding at least a portion of the substrate from the light source using a reconfigurable mask.

[0100] According to some embodiments, the hyperspectral profile, or other imaging profile, of the cells of an unknown species may be used to identify the species of the cell. For instance, in some embodiments, the method includes identifying the cells of the sample using the hyperspectral profile of the cells. In some embodiments, identifying one or more cells comprises determining a species of the one or more cells. Accordingly, the method includes determining a species of one or more cells at least in part from the hyperspectral profile of the cells, in some embodiments. In some embodiments, the method further comprises identifying one or more cells of interest in one or more cell growth channels based on a sensed property. In some such embodiments, the sensed property may be used in conjunction with the hyperspectral property of the one or more cells. For example, one or more cells contained within a particular growth trench may be identified as eells of interest, e.g., based on a sensed property using imaging teehniques described above. It should be noted the sensed property may include any property that can be directly or indirectly observed through various microscopy techniques, in accordance with various embodiments. For example, the sensed property may include, but are not limited to, a fluorescence property, a cell phenotype, a cell genotype, extracellular secretions, a marker or taggant associated with the cell (e.g., protein, antibody, etc.), etc. Specific examples of sensed properties include, but are not limited to, a cell morphology or physiology, gene expression, cell growth rate, cellular localization patterns, enzymatic activity, DNA replication and modification, chromosome segregation patterns, metabolic state, cell envelopes, spatial distributions of organelles, intracellular structures, cell-cell interactions, cellular secretions, or any combinations thereof.

[0101] Additionally, in some embodiments, a mask (e.g., an electronically reconfigurable mask) may be employed to selectively shield the cells of interest contained in the one or more cell growth trenches from a source of light (e.g., laser light), for instance, is discussed herein. That is, in some embodiments, the method may include employing and / or configuring a reconfigurable mask over the substrate of the system. In some embodiments, a mask may be employed to selectively shield a portion of a cell growth channel containing cells of interest from the source of light. Thus, for example, the cells of interest contained within a first, masked cell growth channel may be spared while other cells contained within other, non-masked cell growth channels may be killed. In some embodiments, reconfiguring a mask as described above to shield certain cells and expose other cells to a light source may be used to selectively kill cells. Such a method of killing cells, in some embodiments, may be useful in retaining and / or collecting certain cells, for instance, cells of interest that were identified by analyzing a hyperspectral profile thereof, or other imaging profile. In some such embodiments, cells of interest may be rare cells, e.g., that are present in a sample in a low number relative to other cells in the same. Further description regarding systems and methods of selectively killing, retaining, and / or collecting cells are described in U.S. Patent Application Serial No. 18 / 140,320, filed April 27, 2023, entitled “Systems and Methods for Retrieving Cells from a Continuous Culture Microfluidic Device.” As mentioned above, in some embodiments, the cells may be present on any of a variety of substrates. Additionally, in some such embodiments, because a hyperspectral profile of single cell may be obtained, the cells may not need to grow before acquiring a hyperspectral profile thereof and identifying one or more of the cells. Thus, in some embodiments, the methods described herein for identifying one or more cells is advantageously fast, for instance, compared to typical cell screening methods. In some embodiments, the time to identify a species of a cell having an unknown species is less than or equal to 24 hours, less than or equal to 18 hours, less than or equal to 12 hours, less than or equal to 6 hours, or less than or equal to 1 hour.

[0102] In one set of embodiments, the system may be employed to detect cell growth in, reactions to, and / or propagation in different environments. Many cellular behaviors depend on growth conditions and / or environmental conditions. The system may allow for multigenerational imaging under many different environments, which allows the monitoring of changes associated with cells and cellular processes between environments. A large number of genetic variants can be monitored in parallel, and cells for variants of interest can then be extracted. In accordance with some embodiments, the growth rate of certain cells may vary based on the growth conditions, which may further be useful in identify a species of a cell. In some embodiments, a method may include growing a plurality of cells within a plurality of cell growth trenches under a first set of conditions and acquiring a first hyperspectral profile of at least one of the cells within at least one of the cell growth trenches. In some such embodiments, following this growth, the method may further include growing the plurality of cells within the plurality of cell growth trenches under a second set of conditions and acquiring a second hyperspectral profile of the at least one of the cells within the at least one of the cell growth trenches. In some embodiments, a first hyperspectral profile may be obtained of cells before exposure to a first cell growth medium, with subsequent exposure to the first cell growth medium and acquisition of a second hyperspectral profile. Similarly, in some such embodiments, the cells may then be exposed to a second cell growth medium, and a third hyperspectral profile may be acquired after providing time for the cells to grow. Exposure to various cell growth media (e.g., a third cell growth medium, a fourth cell growth medium, and so forth) followed by corresponding hyperspectral imaging may be continued with any suitable number of cell growth media to determine cell growth rates or other features of the cells, in accordance with some embodiments. Any of a variety of conditions are suitable for exposing cells to determine reactions, growth, or the like, in accordance with some embodiments. For instance, in some embodiments, the conditions may include different solution environments, including various buffers, electrolyte compositions, pharmaceutical compounds, antibiotics, experimental compounds, or the like. Similarly, the lighting conditions, temperatures, pH values, or the like of the environment may be varied, in accordance with some embodiments. In some embodiments, it may be advantageous to include one or more antibiotic compounds within the cell growth conditions and to observe cellular responses to identify certain bacteria within a population of cells. For instance, in some cases, a population of cells may be imaged at a first time before exposure to the antibiotic compound using hyperspectral imaging, whereafter the population is then exposed to the antibiotic compound for a certain amount of time (e.g., at least 10 s, at least 1 min, at least 10 min, at least 1 hour, or at least 6 hours, and / or no more than 12 hours, no more than 1 day, or no more than 1 week). The population of cells may then again be imaged using hyperspectral imaging at a second time following exposure to the antibiotic compound, according to some embodiments. The hyperspectral images from the first and second time may then be compared to determine cellular growth and / or cellular reactions to the antibiotic compound and to identify an identity of one or more unknown cells within the population of cells, according to some embodiments. It is also possible to use similar experimental arrangements with different compounds to determine cellular responses in different conditions, in some instances, e.g., an experimental compound configured to treat a condition being exposed to a population of cells having the condition.

[0103] The methods described herein may be implemented by one or more controllers including at least one processor operatively coupled to the various controllable portions of a hyperspectral imaging system, or other imaging system, such as disclosed herein. The method may be embodied as computer readable instructions stored on non-transitory computer readable memory associated with the at least one processor such that when executed by the at least one processor the system may perform any of the actions related to the methods disclosed herein. Additionally, it should be understood that the disclosed order of the steps is exemplary and that the disclosed steps may be performed in a different order, simultaneously, and / or may include one or more additional intermediate steps not shown as the disclosure is not so limited. In some embodiments, when executed on by the one or more processes, the computer readable instructions stored on non-transitory computer readable memory may be configured to process data associated with the images (e.g., the hyperspectral images). In some embodiments, processing may include training and / or utilizing one or more models to identify imaged cells using one or more classes of information from the images.

[0104] In some embodiments, the processing comprises training and / or using a Convolutional Neural Network. However, other Al-driven systems for applications such as cell determination, etc., as discussed herein, are also possible in other embodiments. In some cases, the Al-driven systems may comprise a computer-readable medium configured to process data transmitted and / or received (e.g., hyperspectral data, etc.) using convolutional neural networks, recurrent neural networks, Gaussian and / or Bayesian processes, artificial neural networks, generative adversarial networks, and / or other machine learning techniques. Based, at least in part, on the data transmitted and / or received, the computer-readable medium can be used to determine cells as discussed herein.

[0105] International Patent Application No. PCT / US2020 / 038867, filed on June 22, 2020, published as international Patent Publication No. WO 2020 / 257746 on December 24, 2020, and entitled “Isolating Live Cells After High-Throughput, Long-Tenn, Time-Lapse Microscopy,” by Luro, et al., is incorporated herein by reference in its entirety. U.S. Patent Application Serial No. 18 / 140320, filed April 27, 2023, entitled “Systems and Methods for Retrieving Cells from a Continuous Culture Microfluidic Device,” is incorporated herein by reference in its entirety.

[0106] In addition, the following are each incorporated by reference in their entireties: U.S. Provisional Patent Application Serial No. 63 / 646,534, filed May 13, 2024, entitled “Systems and Methods for Identifying Cells Using Hyperspectral Imaging and Other Techniques,” by Paulsson, et al., U.S. Provisional Patent Application Serial No. 63 / 646,582, filed May 13, 2024, entitled “Identifying Bacteria and other Organisms by Determining Growth Profiles,” by Paulsson, et al., and U.S. Provisional Patent Application Serial No. 63 / 646,599, filed May 13, 2024, entitled “Microscopy Systems and Methods for Identifying Bacteria and Other Organisms in Microfluidic Devices,” by Paulsson, el al.

[0107] The following examples are intended to illustrate certain embodiments of the present disclosure, but do not exemplify the full scope of the disclosure. EXAMPLE 1

[0108] The following example described an assay that extracts autofluorescence information from individual bacterial cells to predict an identity thereof.

[0109] The assay relies on imaging the cells with different excitation wavelengths and collecting the emitted light by the cells in different ranges of wavelengths. The position of the cells detected is extracted from a phase contrast image. High quality microfluidic devices ensure low background noise signal. A conceptual view of the images is shown in FIG. 5. From each distinct fluorescence image, information can be extracted that relates back to specific emission bands in the electromagnetic spectrum. By pooling all the images together, the spectral signature that is unique to the amounts, kinds, and positions of molecules in a particular cell at the moment of capture can be composed, as shown in FIG. 6.

[0110] A library consisting of 25 bacterial species was pooled together, 14 of the species were different strains of the same Escherichia coli, each strain expressing a different encapsulin protein under a constitutive promoter. Three of the species were different clinical isolates of Acinetobacter baumanii, a cyanobacteria Synechococcus elongatus, Aeromonas veronii, Bacillus thurigiensis, Leuconostoc pseudomesenteroides, Pseudomonas azotoformans, Sphingobacter mizutaii and Streptomyces toxytricini. With the extended bacteria pool of 25 species, the model was able to predict identity of unknown species. Any significant error remained within clusters of related bacteria, either the same species or same morphology.

[0111] Looking closer at the E. coli cluster in the confusion matrix, it was observed that the highest source of error in the model comes from the fact this cluster comprises roughly 60% of the data. However, the model remarkably retains the identity of each strain in all of the cases, as shown in Figure 7. Upon collapsing that cluster on a single label the predicted power on E. coli becomes 99% as it does for the rest of the species, and it remains high for the rest of the species. Figure 8 shows this in an updated confusion matrix. In FIG. 8, it can also be seen that the predictions for the Acinetobacter sp. remain within the same cluster, which means that in the worst case the species can be distinguished and only the more virulent version of one of the species would be misclassified as the less virulent or non-virulent subspecies of the bacteria with 15% probability. These are bacteria that share close to identical morphological distributions. The difference in the spectral signatures can be seen in FIG. 9, which plots the normalized mean fluorescence intensities for three different isolates of A. baumanii having different trajectories. The two more closely matching spectral signatures within this cluster of 3 are known to have a closer common ancestor than the third one.

[0112] To push the limits of differentiation, along an E. coli growth curve measurements were taken at different timepoints, as shown in FIG. 10, from which it can be seen that the accuracy of the model (trained on cells at stationary phase) stays similar at first, slow drops, and then slowly rises as it approaches the same time that the cells used as training data spent in the culture. This is interesting because it is expected for the cells to be more metabolically similar at the start and end of the growth curve than in the middle.

[0113] Replicates of the data collection and model training were performed. The replicates were performed on pooled samples of 35 and 30 members, the results of which are shown in FIGS. 11 and 12, respectively. In each case, the model retained high accuracy for identifying the bacterial cell species.

[0114] There are various advantages of the system. For example, little sample preparation is required, as there is no need to condition the cells and they can be imaged as they are or can be incorporated into microfluidics routines. Moreover, the images are captured using a microscope which allows for accurate direct measurements of morphological information of the cells, e.g., to correlate with the hyperspectral profile). Advantageously, the technique is fast, as the total time from sample preparation to prediction of identity for a single cell (field of view) is 3 minutes for sample preparation, 33 seconds of imaging, on the order of ms of image analysis, and prediction. This makes the technique faster than sequencing, which is the fastest of typical cell screening techniques as it requires the least amount of sample (i.e., less time is needed for cell growth and replication before identifying). Desirably, this technique can detect very rare members of the population, since each and every one of the cells in a given sample, may be detected. This technique doesn’t rely on amplification to pick up a signal. That is, if a particular- cell is present in a sample, it is likely to be detected. It should be understood, however, that in some cases, certain cells may not be detected if the sample is large enough and / or the rare cell is present in a low concentration.

[0115] EXAMPLE 2 In this example, separating cells of a sample in a microfluidic device is described.

[0116] To address cells (e.g., bacterial cells) collected from natural settings / clinical samples / etc., which may not be in the same conditions and / or are incomparable to the cells used to form the library, the assay was performed in a microfluidic device having a plurality of cell growth trenches. Within the microfluidic device, the cells can be loaded and grown for a certain period of time under any of a variety of conditions, which are all compatible with the hyperspectral imaging of Example 1. Using this method, cells collected from a natural setting, a clinical sample, etc. may be brought to a comparable field (e.g., under similar conditions), and may also be analyzed under changing conditions to facilitate multiple comparable fields from which to extract fluorescence signatures.

[0117] Additionally, because of the geometry of this microfluidic device, multiple copies of the single cells having an unknown species may be obtained, and thus the signal can then be processed computationally and the error in the average of the signal may be decreased.

[0118] EXAMPLE 3

[0119] In this example, hyperspectral profiles of bacteria cells were acquired and used to classify (e.g., identify a species) cells.

[0120] FIG. 13A shows a schematic illustration of a plurality of cells contained within cell growth trenches of a microfluidic device. Light having an excitation wavelength was exposed to the ells along the grayed line in the x-direction, perpendicular to the cell growth trenches. The excitation wavelengths of the light, in this Example, were 375 nm, 440 nm, 475 nm, 510 nm, 555 nm, and 575 nm. FIG. 13B is a schematic illustration depicting that the excitation wavelength is reflected towards the cells via a mirror (e.g., an optical mirror), whereafter autofluorescence from the cells is allowed to pass through the mirror toward a prism. The prism splits the diffracts the light into its constituent wavelengths to be measured at the detector to obtain the corresponding hyperspectral profile. FIG. 13C shows an image of autofluorescence from different cells in adjacent cell growth trenches as a function of different excitation wavelengths (e.g., 440 nm, 475 nm, and 555 nm).

[0121] FIG. 14A shows hyperspectral profiles obtained for eight distinct bacteria species contained within microfluidic devices using the hyperspectral imaging system described in FIGS. 13A-13C. FIG. 14B shows the confusion matric corresponding to the hyperspectral profiles obtained from the bacteria in FIG. 14A. The accuracy of identifying the bacteria using the hyperspectral profiles was greater than or equal to 0.96 for all the tested species, with accuracies up to 1.

[0122] EXAMPLE 4

[0123] This example describes processes through which hyperspectral images may be processed and analyzed to identify cells.

[0124] The described system processes input data, initially provided as multi-channel two- dimensional arrays (X) with corresponding labels (y), retrieved iteratively from distinct data files. The system first systematically transforms the input data X by reshaping it into a predefined format comprising specific channel, height, and width dimensions (e.g., 25, 64, 64), 24 "spectral channels" one phase contrast channel, images of 64 by 64. The associated labels y are converted from their original format into a numerical integer representation using a label encoding process, and subsequently transformed into a binary vector format (one-hot encoding) suitable for multi-class classification.

[0125] Within each iteration of processing a data segment, the system randomly shuffles the data arrays and their corresponding transformed labels. This shuffled segment is then partitioned into two subsets: a primary subset (e.g., 80%) designated for model training, and a secondary subset (20%) reserved for validation. Furthermore, the numerical values within the input data arrays X in both subsets are scaled via division by a constant factor (e.g., 65535, camera bit depth) to noimalize their range. This entire sequence of operations - loading, reshaping, label transformation, shuffling, partitioning, and normalization - is repeated iteratively for multiple distinct data segments, with the resulting processed training and validation subsets being sequentially fed into a computational model (specifically, a Convolutional Neural Network in this example) for incremental training and concurrent performance evaluation.

[0126] The computational core of the system comprises a specific type of predictive model known as a Convolutional Neural Network (CNN), constructed as a defined, sequential arrangement of processing layers. This network is configured to receive input data structured as a multi-dimensional array with predefined dimensions (i.e., 25 channels and 64x64 spatial dimensions in this example). Following the input, the data propagates through a series of convolutional processing layers, each applying a consistent number of 256 filters and kernel of size 3x3 and a non-linear activation function to progressively extract hierarchical features. Interspersed within this sequence are pooling layers and dropout layers designed to systematically reduce the spatial dimensions of the intermediate data representations and as a regularization mechanism. Subsequent to the feature extraction and pooling stages, a flattening layer transforms the processed multi-dimensional data into a one-dimensional vector format. This vector is then processed by a final fully connected output layer which utilizes a softmax activation function to generate a probability distribution across a set of predefined output classes, thereby providing the final classification determination

[0127] The exact number of layers was tuned to the classification task, because the number of distinct output classes to be predicted is a key factor influencing the necessary model complexity. For instance, a classification task requiring the differentiation between a smaller set of categories (e.g., 7 classes as in this ESKAPEE example) may be effectively addressed by a model configuration utilizing comparatively fewer parameters than a task demanding prediction across a significantly larger set of categories (e.g., 30 or 35 classes).

[0128] This is the basis for the confusion matrices produced with the above-noted kind of data. With that data set of already collected pictures, the following images were aggregated to produce two more models. One of ESKAPEE and one of the same microbes but the training data was augmented with related species that were called ESKAPEE extended universe. For example, in the ESKAPEE only Enterococcus faecium (for the E in ESKAPEE) was included, whereas in the ESKAPEE extended universe, other enterococcus (Enterococcus faecalis, and Enterococcus malodoratus) were also included. The full list of species for this set of models is below:

[0129] FIGS. 15A and 15B show the confusion matrices obtained from the ESKAPEE and the ESKAPEE extended universe sets, respectively. Using only 7 classes during identification, the species in the ESKAPEE sample of FIG. 15A were identified with an accuracy of at least 0.88. Similarly, the species in the ESKAPEE extended universe sample of FIG. 15B were identified with an accuracy of at least 0.81 using only 7 classes during identification. EXAMPLE 5In this example, hyperspectral imaging and fluorescence measurements of bacterial cells were acquired in the presence and absence of a compound configured to enhance the autofluorescence of the bacterial cells.

[0130] Two experiments were performed in this example. In the first experiment, eight bacterial strains were cultured overnight in the presence or absence of 4 mM of 5-aminolevulinic acid (5- ALA). The following day, the cells were immobilized on agar pads and imaged using phase contrast and fluorescence microscopy. Fluorescence imaging was performed with an excitation wavelength of 405 nm and 630 nm emission filters, a 500 ms exposure time, and 100% light source power. Individual bacterial cells were segmented, and per-pixel fluorescence intensity values were extracted for each cell in both 5-ALA-treated and untreated conditions. These single-cell fluorescence measurements were subsequently represented as intensity distributions. FIGS. 16A-16H show the fluorescence measurements for the eight different bacterial strains grown in the presence and absence of the 5-ALA. When grown in the presence of the 5-ALA, the bacterial strains exhibited higher fluorescence counts when compared to when they were grown in the absence of the 5-ALA. This data shows that the 5-ALA-induced increase in autofluorescence can be detected at the single-cell level in all the tested strains except in Enterococcus faecalis (which has an incomplete heme biosynthetic pathway). The increase in autofluorescence highlights differences between different bacteria and facilitates improved hyperspectral bacterial identification.

[0131] In the second experiment, E. coli cells were loaded into a microfluidic device comprising cell growth trenches. The cells were then exposed to one of the following conditions: unsupplemented growth media (i.e., “media” in FIG. 17), media supplemented with 4 mM of 5- aminolevulinic acid (i.e., “media+5-ALA” in FIG. 17), or buffer supplemented with 4 mM of 5- aminolevulinic acid (i.e., “Buffer+5-ALA” in FIG. 17). After 30 min incubation, the cells exposed to 5-ALA either in media or buffer, became visibly fluorescent as shown in FIG. 17. Fluorescence imaging was performed with 405 nm excitation wavelength and 630 nm emission filters, a 500 ms exposure time, and 100% light source power. This data further shows that the 5- ALA-induced increase in autofluorescence can be detected at the single-cell level in microfluidic devices.

[0132] EXAMPLE 6 In this example, experimental considerations for hyperspectral imaging of cells are described.

[0133] FIGS. 18A-18B shows a schematic diagram detailing how hyperspectral imaging of a cell is performed with structured illumination. FIG. 18 shows how structured illumination can be used to encode spectral information into the image by illuminating with a thin line avoids spectra from different cells that could overlap spatially. The light is incident on the device between the dashed lines. FIG. 18C shows an image of a portion of the microfluidic device containing TetraSpeck microspheres, and a corresponding hyperspectral image of the microspheres. FIG. 18D shows the brightfield image (top half) of a microfluidic device comprising cells in cell growth trenches, with dashed lines indicating where light is incident during hyperspectral imaging. Structured illumination is used to illuminate the device between the dashed lines. FIG. 18D further shows the corresponding hyperspectral profiles obtained from the microfluidic device (bottom half), and a hyperspectral profile plotted for one location in the microfluidic device, where with emission spectra using different excitation wavelengths arc plotted.

[0134] FIG. 19 shows an example where a microlens array is used to generate multifocal excitation for hyperspectral imaging. By passing light through a prism in the emission path, each excitation spot is decomposed in a local spectrum. By scanning this grid during a single camera exposure, instant structured illumination microscopy (iSIM) can be performed or scanning and imaging sequentially for hyperspectral imaging.

[0135] Specifically, in this arrangement, an alternative light path was added in the emission path to incorporate a dispersive optical element such as the prism. In this manner, the light from the emission spots that are produced when exciting the sample, are spectrally decomposed when they pass through the prism. This way each spot now becomes a line representing the local spectrum contained within that point. Unlike imaging where the scanning of the excitation is done during a single camera exposure, in hyperspectral mode, an image is acquired at each position of the scanned excitation. In this implementation, the prism was placed in between the alternative relay system (within the alternative light path) between relay lenses. However, the prism could also be contained anywhere in the traditional emission path including the conjugate image and Fourier spaces. With this implementation, by changing between the traditional and alternative emission paths the microscope can switch between standard super-resolution fluorescence imaging and hyperspectral imaging. Other ways to include the prism could be using a flip mount to reversibly insert and remove the prism from the emission light path, or by using a pair of Amici prisms where rotating the relative angle between the two prisms could tune the dispersion of those collective elements to switch from imaging to hyperspectral modes. Based on the insight provided by this disclosure, other experimental designs may be considered to achieve iSIM and / or scanning and imaging sequentially for hyperspectral imaging.

[0136] FIG. 20A shows how considerations to improve the hyperspectral multifocal pattern for encoding the hyperspectral information into the sparse regions of the illumination pattern, as well as a demonstration of hyperspectral iSIM. The data show how excitation spots are spectrally decomposed to plot different wavelengths in different positions along the axis of dispersion. The data was acquired by shining light through different emission filters corresponding to the Texas Red, GFP, and DAPI filters.

[0137] FIG. 20B-20C show examples of different ways to calibrate the recorded spectra. FIG. 20B shows Off-sample target wavelength where an LED dome is used to transmit in combination with a narrow emission notch filter to detect a set wavelength. For example, light of a known wavelength can be shined through the hyperspectral detection light path to mark specific wavelengths in the recorded spectra. FIG. 20C is on-sample target wavelength using known autofluorescence (e.g., from PDMS or other samples) to calibrate and align other recorded spectra. Doping materials in the device, e.g., with fluorescent beads, quantum dots, dyes, etc., may also provide a known autofluorescence profile to calibrate with. Note that inserting a notch filter to reject a specific wavelength from the recorded spectra can also be used to calibrate and align the recorded spectra.

[0138] EXAMPLE 7

[0139] In this example, hyperspectral imaging and fluorescence measurements of bacterial cells were acquired in the various media.

[0140] Cells in different growth conditions and in the presence or absence of the autofluorescence-enhancing compound 5-ALA were imaged using hyperspectral imaging. Specifically, the cells were either grown in stationary or exponential cell media. The cell species and corresponding IDs are noted below in Table 1.

[0141] Table 1. Cell species and IDs.

[0142] FIGS. 21A-21D show the raw data for identification of the cell species in different growth media. In FIG. 21 A, the growth medium was exponential and 5-ALA was present, and the hyperspectral imaging accuracy was 89.4%. In FIG. 21B, the growth medium was stationary and 5-ALA was present, and the hyperspectral imaging accuracy was 87.0%. In FIG. 21C, the growth medium was stationary and 5-ALA was absent, and the hyperspectral imaging accuracy was 83.5%. In FIG. 21D, the growth medium was exponential and 5-ALA was absent, and the hyperspectral imaging accuracy was 88.3%. These data are demonstrate high accuracy identification of the bacterial species using hyperspectral imaging. Additionally, the data indicate the presence of the 5-ALA and the use of exponential growth media may increase accuracy of bacterial identification using hyperspectral imaging. FIGS. 22A-22D show the confusion matrices corresponding to the data shown in FIGS. 21A-21D.

[0143] EXAMPLE 8

[0144] In this example, hyperspectral imaging for antibiotic susceptibility testing was demonstrated.

[0145] Staphylococcus aureus MRSA USA300 cells were loaded into a microfluidic device and allowed to grow in medium for 1-2 hours. The cell medium was then switched to a non- autofluorescent MOPS buffer to collect baseline hyperspectral profiles at three excitation wavelengths (374 nm, 456 nm, and 486 nm) across different fields of view for individual cells. After establishing baseline measurements, MOPS buffer was then supplemented with the Gentamicin (3.75 pg / mL, 15x MIC). FIGS. 23A-23D shows hyperspectral profiles obtained for a 1 minute baseline, 1 minute after antibiotic treatment, for a 3 minute baseline, and 3 minutes after antibiotic treatment. Similar data were collected for 5 minute, 7 minute, 10 minute, 15 minute, 30 minute, and 1 hour timelines. Photobleaching was controlled for. FIG. 24 are plots showing the difference in the intensity of the hyperspectral profiles obtained at each time (e.g., final intensity - initial intensity).

[0146] The experiments were repeated using E. coli MG 1655 AmotA cells and a corresponding Erythromycin (250 pg / mL, approximately 5x MIC) antibiotic. Data were collected at the same timepoints as noted above for the previous experiment in this example. FIG. 25 shows plots of the difference in the intensity of the hyperspectral profiles obtained at each time (e.g., final intensity - initial intensity). The observable differences shown in FIGS. 24-25 indicate that hyperspectral imaging can be used to test the response of bacterial species to antibiotics.

[0147] While several embodiments of the present disclosure have been described and illustrated herein, those of ordinary skill in the art will readily envision a variety of other means and / or structures for performing the functions and / or obtaining the results and / or one or more of the advantages described herein, and each of such variations and / or modifications is deemed to be within the scope of the present disclosure. More generally, those skilled in the art will readily appreciate that all parameters, dimensions, materials, and configurations described herein are meant to be exemplary and that the actual parameters, dimensions, materials, and / or configurations will depend upon the specific application or applications for which the teachings of the present disclosure is / are used. Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the disclosure described herein. It is, therefore, to be understood that the foregoing embodiments are presented by way of example only and that, within the scope of the appended claims and equivalents thereto, the disclosure may be practiced otherwise than as specifically described and claimed. The present disclosure is directed to each individual feature, system, article, material, kit, and / or method described herein. In addition, any combination of two or more such features, systems, articles, materials, kits, and / or methods, if such features, systems, articles, materials, kits, and / or methods are not mutually inconsistent, is included within the scope of the present disclosure.

[0148] In cases where the present specification and a document incorporated by reference include conflicting and / or inconsistent disclosure, the present specification shall control. If two or more documents incorporated by reference include conflicting and / or inconsistent disclosure with respect to each other, then the document having the later effective date shall control.

[0149] All definitions, as defined and used herein, should be understood to control over dictionary definitions, definitions in documents incorporated by reference, and / or ordinary meanings of the defined terms.

[0150] The indefinite articles “a” and “an,” as used herein in the specification and in the claims, unless clearly indicated to the contrary, should be understood to mean “at least one.”

[0151] The phrase “and / or,” as used herein in the specification and in the claims, should be understood to mean “either or both” of the elements so conjoined, i.e., elements that are conjunctively present in some cases and disjunctively present in other cases. Multiple elements listed with “and / or” should be construed in the same fashion, i.e., “one or more” of the elements so conjoined. Other elements may optionally be present other than the elements specifically identified by the “and / or” clause, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, a reference to “A and / or B”, when used in conjunction with open-ended language such as “comprising” can refer, in one embodiment, to A only (optionally including elements other than B); in another embodiment, to B only (optionally including elements other than A); in yet another embodiment, to both A and B (optionally including other elements); etc.

[0152] As used herein in the specification and in the claims, “or” should be understood to have the same meaning as “and / or” as defined above. For example, when separating items in a list, “or” or “and / or” shall be interpreted as being inclusive, i.e., the inclusion of at least one, but also including more than one, of a number or list of elements, and, optionally, additional unlisted items. Only terms clearly indicated to the contrary, such as “only one of’ or “exactly one of,” or, when used in the claims, “consisting of,” will refer to the inclusion of exactly one element of a number or list of elements. In general, the term “or” as used herein shall only be interpreted as indicating exclusive alternatives (i.e. “one or the other but not both”) when preceded by terms of exclusivity, such as “either,” “one of,” “only one of,” or “exactly one of.”

[0153] As used herein in the specification and in the claims, the phrase “at least one,” in reference to a list of one or more elements, should be understood to mean at least one element selected from any one or more of the elements in the list of elements, but not necessarily including at least one of each and every element specifically listed within the list of elements and not excluding any combinations of elements in the list of elements. This definition also allows that elements may optionally be present other than the elements specifically identified within the list of elements to which the phrase “at least one” refers, whether related or unrelated to those elements specifically identified. Thus, as a non-limiting example, “at least one of A and B” (or, equivalently, “at least one of A or B,” or, equivalently “at least one of A and / or B”) can refer, in one embodiment, to at least one, optionally including more than one, A, with no B present (and optionally including elements other than B); in another embodiment, to at least one, optionally including more than one, B, with no A present (and optionally including elements other than A); in yet another embodiment, to at least one, optionally including more than one, A, and at least one, optionally including more than one, B (and optionally including other elements); etc.

[0154] When the word “about” is used herein in reference to a number, it should be understood that still another embodiment of the disclosure includes that number not modified by the presence of the word “about.”

[0155] It should also be understood that, unless clearly indicated to the contrary, in any methods claimed herein that include more than one step or act, the order of the steps or acts of the method is not necessarily limited to the order in which the steps or acts of the method are recited.

[0156] In the claims, as well as in the specification above, all transitional phrases such as “comprising,” “including,” “carrying,” “having,” “containing,” “involving,” “holding,” “composed of,” and the like are to be understood to be open-ended, i.e., to mean including but not limited to. Only the transitional phrases “consisting of’ and “consisting essentially of’ shall be closed or semi-closed transitional phrases, respectively, as set forth in the United States Patent Office Manual of Patent Examining Procedures, Section 2111.03.

Claims

1. CLAIMSWhat is claimed is:

1. A method, comprising: acquiring hyperspectral profiles of at least some of a plurality of cells within a plurality of cell growth trenches of a microfluidic device.

2. A method, comprising: growing a plurality of cells within a plurality of cell growth trenches under a first set of conditions; acquiring a first hyperspectral profile of at least one of the cells within at least one of the cell growth trenches; thereafter, growing the plurality of cells within the plurality of cell growth trenches under a second set of conditions; and acquiring a second hyperspectral profile of the at least one of the cells within the at least one of the cell growth trenches.

3. A method, comprising: acquiring a hyperspectral profile of only a single cell.

4. The method as in claim 3, further comprising determining a species of the single cell at least in part from the hyperspectral profile.

5. The method as in claim 3 or 4, wherein the single cell is one cell of a plurality of cells.

6. The method as in any one of claims 3-5, wherein the single cell is contained within a microfluidic device containing a plurality of cell growth trenches.

7. A method, comprising:containing a plurality of cells from a sample of one or more different unknown species of unicellular organisms within a microfluidic device; and identifying a species of the unicellular organisms within the microfluidic device in less than or equal to 1 hour after containing the cells within the microfluidic device.

8. The method as in any one of the preceding claims, wherein the plurality of cells comprises a plurality of bacteria cells.

9. The method as in any one of the preceding claims, further comprising introducing the plurality of cells into a microfluidic device comprising one or more cell growth trenches.

10. The method as in any one of the preceding claims, further comprising growing the plurality of cells under a first condition and acquiring a first hyperspectral profile of at least some of the plurality of cells.

11. The method as in claim 10, following growing the cells under the first condition and acquiring the first hyperspectral profile, further comprising growing the cells into a microfluidic device comprising one or more cell growth trenches under a second condition and acquiring a second hyperspectral profile of at least some of the plurality of cells.

12. The method of any one of the preceding claims, further comprising providing the plurality of cells.

13. The method of claim 12, wherein providing the plurality of cells comprising acquiring the cells by sampling cells from a source.

14. The method of claim 13, wherein the source comprises a body of water.

15. The method of claim 13, wherein the source comprises soil.

16. The method of claim 13, wherein the source comprises a food source.

17. The method of claim 13, wherein the source comprises a subject having a disease.

18. A system, comprising: a microfluidic device, comprising: a cell flow layer comprising a growth channel having an inlet portion, an outlet portion, an inlet valve portion associated with the inlet portion, and an outlet valve portion associated with the outlet portion; a plurality of cell growth trenches configured to contain cells, the plurality of cell growth trenches being fluidically coupled to the growth channel; and a control layer configured to control flow of fluid in the cell flow layer; and a light source configured for fluorescence microscopy and positioned to direct light at at least a portion of the plurality of cell growth trenches of the microfluidic device; a mask positioned along an optical path between the light source and the device; a detector positioned to receive fluorescence from the cells illuminated by the light; and a processor configured and arranged to perform hyperspectral analysis on the fluorescence received by the detector.

19. The system as in claim 18, wherein the mask is electronically reconfigurable.

20. The system as in claim 18 of 19, wherein the light source is a laser.

21. The system as in any one of claim 18-20, wherein the detector is a CCD.

22. The system as in any one of claims 18-21, wherein the microfluidic device comprises a plurality of cells, some of the plurality of cells are of an unknown species.

23. The system as in any one of claims 18-22, wherein the system is configured to identify a species of a cell having an unknown species contained within the microfluidic device.

Images