Characterizing Cell Health Using Machine Learning

Abstract

A sequence of microscopic images, each comprising an array of pixels, visualizing fluorescently labeled cellular components within live cells is received. These images are segmented to identify one or more cells visualized within the images. A machine learning model assigns a cell health score and / or cell health subtype score to each pixel of a particular cell. The cell health scores and / or cell health subtype scores based on these pixels are used to calculate a total cell health score and / or total cell health subtype score for each cell. Data characterizing the calculated scores is provided to a consuming application or process.

Description

[Technical Field]

[0001] Related Applications This application claims priority to U.S. Patent Application No. 63 / 444,543, filed February 9, 2023, the contents of which are incorporated by reference in their entirety.

[0002] The subject matter described herein relates to machine learning-based techniques for characterizing the health of cells (e.g., living cells) from images generated by a microscopy system (e.g., a super-resolution microscopy system). [Background technology]

[0003] Protein movement within the dense environment of living cells is strongly influenced by interactions with their surroundings. Single-molecule tracking (SMT) is one method for capturing protein movement as a reporter of activity. SMT involves imaging fluorescent proteins of interest with high spatiotemporal resolution to track their movement within complex systems, such as living cells. The information embedded in these traces has been used to investigate diverse cellular phenomena, including protein-protein interactions, such as those mediating signal transduction, interorganelle communication, nuclear organization, and transcriptional regulation. SMT also enables the analysis of cells while they are perturbed by various compounds and environmental conditions. Summary of the Invention

[0004] In a first aspect, a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received. Each of these images includes an array of pixels. Each image is segmented to identify one or more cells. Each pixel of each identified cell is then assigned a cell health score using at least one machine learning model. For each identified image, a total cell health score can be calculated based on the cell health scores assigned to the pixels. Data characterizing the calculated total cell health score can be provided to a consuming application or process. Related provision can include displaying the score, loading the score into memory of a computing device, storing the score in a physically persistent state, and / or transmitting the score over a network to a remote computing system.

[0005] Microscopic image sequences can visualize fluorescently labeled cellular components while live cells are exposed to compounds or various environmental conditions. Microscopic images can be at the field level. In such variations, the percentage of healthy cells within the field can be calculated (in addition to or as an alternative metric for individual cells).

[0006] Live cells can be stained using a variety of labeling dyes, including Hoechst dyes.

[0007] At least one machine model can be trained using a training dataset. Such a training dataset, in some embodiments, can be generated by selectively exposing a sample of live cells to at least one compound at various concentrations. The compound can take various forms, including, but not limited to, staurosporine and / or sorbitol. The training dataset can also be augmented or otherwise enhanced by randomly varying the brightness levels of the images generated for the training dataset to reflect various environmental conditions and / or by including manually labeled images (either at the field level or pixel level). In some variations, this manual annotation can characterize each pixel as corresponding to one of a healthy cell, an unhealthy cell, or background.

[0008] In one variation, the training data set includes: staining a sample of live cells with a fluorescent dye; exposing a first subset of the sample of live cells to a concentration of a first compound at a first concentration corresponding to unhealthy cells, where the first compound affects cell health; exposing a second subset of the sample of live cells to a concentration of the first compound at a second concentration corresponding to healthy cells; and exposing a third subset of the sample of live cells to a concentration of a second compound at a third concentration corresponding to unhealthy cells, where the second compound affects cell health; exposing a fourth subset of the sample of live cells to a concentration of a second compound at a fourth concentration corresponding to healthy cells, generating images for each of the first, second, third, and fourth subsets of the sample of cells using a microscope, labeling the images of the first and third subsets of the sample of live cells as corresponding to unhealthy cells, and labeling the images of the second and fourth subsets of the sample of live cells as corresponding to healthy cells. The training dataset can be enhanced through other actions, such as varying the exposure time of the first and second subsets of the sample of live cells to the first compound and / or varying the exposure time of the third and fourth subsets of the sample of live cells to the second compound.

[0009] The machine learning model(s) utilized herein can take a variety of forms, including, but not limited to, pixel-level convolutional neural networks such as those provided by the U-Net architecture. The machine learning model(s) can be optimized using a variety of loss functions, including weighted categorical cross-entropy.

[0010] In a related aspect, a sequence of microscopic images is received that visualizes fluorescently labeled cellular components within live cells and includes an array of pixels, each corresponding to a specific cell. Each image is segmented to identify one or more cells. Each pixel is then classified as corresponding to a healthy cell, an unhealthy cell, or background using at least one machine learning model. For each identified image, a total cell health score can be calculated based on the classification of the corresponding pixels. Such a total cell health score can be provided to a consuming application or process.

[0011] In a further related aspect, a sequence of microscopic images is received that visualizes fluorescently labeled cellular components within live cells and includes an array of pixels, each corresponding to a specific cell. Each image is segmented to identify one or more cells. Each pixel is then classified as corresponding to a healthy or unhealthy cell using at least one machine learning model trained using weak supervision, in which training images are labeled at the field level (as opposed to the pixel level). For each identified image, an overall cell health score can be calculated based on the classification of the corresponding pixels. Such an overall cell health score can be provided to a consuming application or process.

[0012] In yet a further interrelated aspect, a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received. Each of these images includes an array of pixels. Each image is segmented to identify one or more cells. Each pixel of each identified cell is then assigned a cell health subtype score using at least one machine learning model. A total cell health subtype score can be calculated for each identified image based on the cell health subtype scores assigned to the pixels. Data characterizing the calculated total cell health subtype score can be provided to a consuming application or process.

[0013] In another related aspect, a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received. Each image is segmented to identify one or more cells. A cell health score and at least one cell health subtype score are determined for each identified cell using an ensemble of separately trained machine learning models. Data characterizing the scores can be provided to a consuming application or process.

[0014] Non-transitory computer program products (i.e., tangibly embodied computer program products) that store instructions, which, when executed by one or more data processors of one or more computing systems, cause at least one data processor to perform the operations described herein, are also described. Similarly, computer systems are described that may include one or more data processors and memory coupled to the one or more data processors. The memory may temporarily or permanently store instructions that cause at least one processor to perform one or more of the operations described herein. Furthermore, methods may be performed by one or more data processors, whether within a single computing system or distributed across two or more computing systems. Such computing systems may be connected via one or more connections, including, but not limited to, connections via a network (e.g., the Internet, a wireless wide area network, a local area network, a wide area network, a wired network, etc.), direct connections between one or more of the computing systems, etc., and may exchange data and / or commands or other instructions, etc.

[0015] The subject matter described herein offers many technical advantages. For example, the present subject matter provides enhanced machine learning-based techniques for characterizing cellular health from images generated by ultra-resolution or other high-resolution microscopy systems. In particular, the characterization provided by the present subject matter increases the likelihood that biological measurements are performed under physiologically relevant conditions with minimal confounding factors.

[0016] The details of one or more variations of the subject matter described herein are set forth in the accompanying drawings and the description below. Other features and advantages of the subject matter described herein will be apparent from the description and drawings, and from the claims.

[0017] This patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee. [Brief explanation of the drawings]

[0018] [Figure 1] A schematic diagram of the htSMT workflow is shown.

[0019] [Figure 2A] 1 illustrates an exemplary image acquisition system of the present disclosure, with the XZ sample plane visible. [Figure 2B] 1 illustrates an exemplary image acquisition system of the present disclosure, with the YZ sample plane visible. [Figure 2C] 1 illustrates an exemplary image acquisition system of the present disclosure, with the XZ sample plane visible. [Figure 2D] 1 illustrates an exemplary image acquisition system of the present disclosure, with the YZ sample plane visible. [Figure 2E] Details of the light beam compared to the HILO-based approach are shown. [Figure 2F] An example of incorporating a camera roll shutter is shown below.

[0020] [Figure 3] 1 shows a schematic diagram of an exemplary sample handling system of the present disclosure.

[0021] [Figure 4] An exemplary system for a high-throughput single-molecule imaging platform for measuring protein movement in living cells is shown.

[0022] [Figure 5] FIG. 1 illustrates data flow through an exemplary system for a high-throughput single-molecule imaging platform that measures protein movement in living cells.

[0023] [Figure 6] 10A-10C are images showing the differences between mask categories and between instance / semantic masks.

[0024] [Figure 7] 1 illustrates an exemplary computer-implemented environment relevant to the subject matter described herein.

[0025] [Figure 8] FIG. 1 illustrates a sample computing device architecture for implementing various aspects described herein.

[0026] [Figure 9] FIG. 1 is a pictorial representation of healthy and unhealthy cells.

[0027] [Figure 10] Shown are microscopic images of a healthy sample, a sample exposed to a first compound (sorbitol), and a sample exposed to a second compound (staurosporine).

[0028] [Figure 11] 1 is an exemplary machine learning architecture for characterizing cell health from microscopic images.

[0029] [Figure 12] 10 shows augmented training data images with selectively varying brightness levels.

[0030] [Figure 13] This shows the classification of an image after the machine learning model has been trained in a weakly supervised manner.

[0031] [Figure 14] FIG. 1 illustrates image classification by a machine learning model enhanced with manually annotated results.

[0032] [Figure 15] Figure 1 shows cell health scores when a mixed group of engineered cell lines with G418 resistance are subjected to G418 treatment.

[0033] [Figure 16] 1 shows the correlation between MPDC and cell health scores upon treatment with topoisomerase inhibitors.

[0034] [Figure 17] The images corresponding to the same processing are shown in FIG.

[0035] [Figure 18] Two sets of input images and the resulting cell health and apoptosis scores are shown.

[0036] [Figure 19] FIG. 1 is a process flow diagram for characterizing cell health (e.g., providing a score) using machine learning.

[0037] [Figure 20] FIG. 1 is a process flow diagram for characterizing (e.g., providing a score for) cell health subtypes using machine learning.

[0038] [Figure 21] FIG. 1 is a process flow diagram for characterizing cell health and cell health subtypes using an ensemble of machine learning models. DETAILED DESCRIPTION OF THE INVENTION

[0039] The presently disclosed subject matter is directed to machine learning-based techniques for characterizing cell health (including cell health subtypes) in microscopy images (e.g., ultra-resolution and other high-resolution microscopy system images). In particular, the present subject matter provides a machine learning model that identifies cell health as a function of compound treatment in Hoechst-labeled fluorescent microscopy images. As described in further detail below, a training set was generated by exposing U2OS cells to various concentrations of staurosporine and sorbitol for various compound incubation times. A weakly supervised training method was used to train a machine learning model (e.g., a U-Net architecture model) to provide a pixel-level binary classification describing whether a pixel belongs to an unhealthy cell or not. The model(s) take as input a single two-dimensional Hoechst image, which has been experimentally demonstrated to successfully capture several different subtypes that can visually manifest as unhealthy. These subtypes may include, but are not limited to, apoptosis, oxidative stress, and inflammation. During high-throughput screening (HTS), the model can work in conjunction with another segmentation model to label all cells by using the average pixel score per mask instance. In some variations, an ensemble of models can be used to characterize the health of cells.

[0040] Microscopic images can be generated using a variety of ultra-high resolution and high resolution systems, particularly those capable of characterizing the dynamics of living cells / molecules / cellular components over time and in various environmental conditions. One example is a microscopic system that uses oblique line scanning (OLS) illumination, which is described in more detail below as a non-limiting example for generating images / videos. Another exemplary microscopic system utilizes thin-film oblique illumination microscopy (HILO). Other microscopic systems can utilize different illumination strategies, including, but not limited to, total internal reflection fluorescence (TIRF), HIST, or SOLEIL, to generate the images analyzed herein.

[0041] The subject matter of the present disclosure may utilize industrial-scale high-throughput SMT (htSMT) technology using optically isolated stimulators (OLS) illumination, systems incorporating such OLS htSMT technology, hardware and software developments related to such OLS htSMT technology, and methods using such OLS htSMT technology. For example, the OLS htSMT technology described herein is capable of measuring protein movement in millions of cells per day. OLS offers advantages such as improved spatial uniformity of the signal-to-noise ratio (SNR) across the camera chip, improved confocality (less out-of-focus signal), and improved temporal resolution, in addition to the ability to capture a large number of cells per field of view, as shown, for example, in Table 1 (each "+" represents a 2-fold improvement). [Table 1]

[0042] The OLS htSMT technology described herein can be used for a variety of applications, including, but not limited to, drug discovery activities such as screening compound libraries and elucidating structure-activity relationships (SAR). Importantly, the OLS htSMT technology described herein can be used to characterize the contributions of both known and novel pathways to larger molecular assemblies that contain targets, such as protein signaling interaction networks.

[0043] 1 , aspects of the present subject matter can be implemented using an OLS htSMT workflow, which can include various stages, such as (i) sample preparation, including reagent handling, (ii) image acquisition using imaging of the sample to generate a series of images and / or videos, (iii) image analysis, for example, using various analytics, single emitter detection and sub-pixel localization (i.e., "super-resolution imaging"), tracking, computer vision, and machine learning algorithms to process these images and videos, (iv) storage of information extracted from or characterizing or constituting the images and videos (i.e., features, raw images, modified images, etc.), and (v) providing insights using the stored information, including biological interpretations (which can additionally or alternatively be provided using various analytics, tracking, computer vision, and machine learning algorithms), as described in more detail below.

[0044] The subject matter of the present disclosure will be described with reference to the figures, wherein reference numerals are used to denote like or equivalent elements throughout. The figures are not drawn to scale and are provided solely to illustrate aspects disclosed herein. Certain disclosed aspects are described below with reference to illustrative example hardware, software, and applications. It should be understood that numerous specific details, relationships, and methods are set forth to provide a more thorough understanding of the subject matter disclosed herein. For clarity of disclosure, and not for purposes of limitation, the detailed description is divided into the following subsections. 1.Definition 2.OLS htSMT hardware 3.OLS htSMT software 4. Specific OLS htSMT Applications 5. Working Example

[0045] 1.Definition Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. In the case of conflict, the present document, including definitions, will control. Preferred methods and materials are described below; however, methods and materials similar or equivalent to those described herein can also be used in the practice or testing of the subject matter of this disclosure. All publications, patent applications, patents, and other references mentioned herein are incorporated herein by reference in their entirety. The materials, methods, and examples disclosed herein are illustrative only and are not intended to be limiting.

[0046] The terms "comprise," "include," "having," "has," "can," "contain," and variations thereof, as used herein, are intended to be open-ended transitional phrases, terms, or phrases that do not exclude the possibility of additional acts or structures. The singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. The present disclosure also contemplates other instances of "comprising," "consisting of," and "consisting essentially of" the instances or elements presented herein, whether explicitly stated or not.

[0047] In reciting numerical ranges herein, each intervening number in the range is expressly contemplated with the same precision. For example, in the range 6 to 9, the numbers 7 and 8 are contemplated in addition to 6 and 9, and in the range 6.0 to 7.0, the numbers 6.0, 6.1, 6.2, 6.3, 6.4, 6.5, 6.6, 6.7, 6.8, 6.9, and 7.0 are expressly contemplated.

[0048] As used herein, the term "about" or "approximately" means within an acceptable error range for a particular value as determined by those skilled in the art, which depends in part on the method of measuring or determining the value, i.e., the limitations of the measurement system. For example, "about" can mean within 3 or more standard deviations, according to the practice in the art. Alternatively, "about" can mean a range of up to 20%, preferably up to 10%, more preferably up to 5%, and even more preferably up to 1% of a given value. Alternatively, particularly with respect to biological systems or processes, this term can mean within an order of magnitude, preferably within 5 times, more preferably within 2 times of a value.

[0049] As used herein, the term "trajectory" refers to a set of spatial coordinates corresponding to the observed positions of fluorescent proteins linked in time. In certain cases, multiple trajectories can be constructed algorithmically by linking multiple fluorescent proteins whose positions have been determined at successive time points. In certain cases, multiple trajectories can be constructed conservatively by linking only spots within a fixed search radius when other links are not valid. In certain cases, multiple trajectories can be constructed probabilistically.

[0050] As defined herein, protein motion refers to changes in the positions of multiple fluorescent proteins. In certain cases, protein motion can be quantified by analyzing changes in spatial coordinates at successive time points. Motion characterized in this way can include, but is not limited to, measuring the distribution of jump lengths. That is, given a set of protein displacements between one time point and a subsequent time point, a histogram of the probability of each displacement length ("jump length") can be constructed. Quantiles of this distribution can be used to describe protein motion. In certain cases, the quantile used is the median of the jump length distribution. In certain cases, the quantile used is the third quartile of the jump length distribution. In certain cases, protein motion can be quantified by analyzing trajectories. Motion characterized in this way can include, but is not limited to, measurements of mean square displacement, defined as the mean value of the squares of all displacements within a trajectory, averaged over multiple trajectories. Motion characterized in this way can also include, but is not limited to, measurements of trajectory length or measurements of the distribution of trajectory lengths. Such characterized motion may include, but is not limited to, measurements of the average radius of gyration, defined by the root-mean-square distance of all coordinates on a trajectory from the center of mass of the set of points included in the trajectory, averaged over multiple trajectories. Such characterized motion may also include, but is not limited to, measurements of the average bond angle, defined by the angle formed from three consecutive spatial coordinates averaged over multiple trajectories. Such characterized motion may also include, but is not limited to, measurements of the maximum likelihood diffusion coefficient estimate, defined as the maximum likelihood diffusion coefficient estimate for multiple trajectories under a single-state diffusion model with a constant localization error. In certain cases, protein motion may be measured through analysis of the products of a link generation algorithm. Such characterized motion may include, but is not limited to, the mean posterior diffusion coefficient, the average of the posterior probability distribution of coefficients from a probabilistic linkage algorithm. Such characterized motion may include, but is not limited to, the geometric mean posterior diffusion coefficient, the average of the log-scaled posterior probability distribution of coefficients from a probabilistic linkage algorithm.In certain cases, protein motion can be measured through model-dependent analysis of multiple trajectories. Motion characterized in this way is the fraction of immobile molecules ("f") defined by two-state model fitting. bound ") may include, but is not limited to:

[0051] As used herein, the term "motion" encompasses not only changes in the direction in which a target moves, but also both increases and decreases in the speed of movement. Thus, tracking motion may, in certain cases, include determining that a target is not moving, e.g., determining that the target is in a statically constrained state or an essentially statically constrained state. Motion can be characterized in various ways, including, but not limited to, quantifying (a) the median of the jump length distribution (the jump length corresponds to the observed distance traveled by the target's fluorescent protein in successive frames), (b) the third quartile of the jump length distribution, (c) the median radius of gyration, (d) the mean posterior diffusion coefficient, (e) the geometric mean posterior diffusion coefficient, (f) the mean square displacement, (g) the median bond angle, (h) the maximum likelihood estimator of the diffusion coefficient, and / or (i) the trajectory length.

[0052] As used herein, detected movement includes, but is not limited to, any change in movement, which may occur in response to any environmental or other factor. For example, but not limited to, movement, or lack thereof, may be caused by (A) the addition of a compound, (B) a change in temperature, (C) a change in oxygen concentration, e.g., the introduction of hypoxia, (D) mechanical stress, (E) a change in pH, and / or (F) a change in light exposure (e.g., an increase or decrease in intensity).

[0053] As used herein, the term "fluorescent protein" refers to any protein that emits a fluorescent signal. In certain cases, the fluorescent emission occurs in response to irradiation with light of a specific wavelength. An example of a naturally occurring fluorescent protein is green fluorescent protein (GFP). However, in certain cases, a protein of interest can be adapted to emit a fluorescent signal through the introduction of an encoded fluorescent tag. That is, a protein sequence is fused to the protein of interest to make it fluorescent. In certain cases, a protein of interest can be adapted to emit a fluorescent signal through the binding of a fluorescent ligand. Non-limiting examples of such encoded fluorescent tags include Halo tag, SNAP tag, CLIP tag, TMP tag, and SunTag. Additionally or alternatively, a protein of interest can be adapted to emit a fluorescent signal by binding to a fluorescent dye molecule, such as an amine-reactive dye or a sulfhydryl-reactive dye.

[0054] As used herein, the term "compound" refers to any chemically defined entity. In certain cases, a compound may be a molecule less than 1000 Da, i.e., a "small molecule." In certain cases, a compound may be a macromolecule, such as a nucleic acid. In certain cases, a nucleic acid may have a defined sequence. In certain cases, a nucleic acid includes (A) ribonucleic acid (RNA) (e.g., including modified RNA), (B) deoxyribonucleic acid (DNA) (e.g., including modified DNA), and (C) a combination of (A) and (B). In certain cases, a nucleic acid is a single- or double-stranded small interfering nucleic acid (e.g., double-stranded siRNA), an antisense oligonucleotide, a ribozyme, a microRNA, or an aptamer. In certain cases, a compound may be a protein. For example, and not by way of limitation, protein compounds of the present disclosure include signaling proteins, such as protein hormones, cytokines, kinases, phosphatases, and other enzymes and transcription factors, as well as antibodies, contractile proteins, structural proteins, storage proteins, and transport proteins. In certain cases, a compound can refer to a mixture of molecules, for example, a mixture of defined composition.

[0055] The term "uniform intensity" as used herein refers to a difference in intensity of no more than 5%, sometimes no more than 10%, or sometimes no more than 15%, relative to the intensity of light, e.g., the intensity of light directed at the sample surface.

[0056] The term "uniform intensity" as used herein, in relation to signal-to-noise ratio (SNR), refers to the SNR on a pixel-by-pixel basis within the field of view (FOV) where the range of possible values ​​falls between 0.5 and 1 standard deviation from the mean SNR.

[0057] 2.OLS htSMT hardware 2.1.Image Acquisition System

[0003] Referring to Figure 1, aspects of the present subject matter can be implemented using an htSMT workflow incorporating a system for image acquisition. For example, such image acquisition can incorporate imaging of a sample to generate a series of images and / or video. Figure 2A shows a schematic diagram of an exemplary image acquisition system of the present disclosure, with the XZ sample plane visible. Figure 2B shows the same exemplary image acquisition system, but with the YZ sample plane visible. An exemplary image acquisition system (2-001) includes a light source (2-005) configured to emit light that is relayed by one or more optical elements in an optical relay (2-010), the optical relay configured to shape the light emitted from the light source to form a shaped beam (2-065) such that the shaped beam has a uniform intensity across the long dimension of the linear shape, the system further including optical elements, such as a galvo mirror (2-085), configured to translate the shaped beam, and one or more optical elements, such as a dichroic mirror (2-100), configured to direct the shaped beam to an objective lens (2-120) so that a portion of the sample surface (2-130) is illuminated by an oblique beam (2-125) resulting in light emission from the sample, such as fluorescent emission, which is focused by the objective lens (2-120) through a series of optical elements, such as a lens (2-155) and an absorption filter (2-160), onto an image collection system (2-165).

[0058] 2.1.1.Light source Referring to the exemplary image acquisition system of FIG. 2A, the system includes a light source (2-005) configured to emit light. The light source (2-005), in certain embodiments of the image acquisition systems disclosed herein, can be configured to emit light at a single wavelength. In certain embodiments of the image acquisition systems disclosed herein, the light source (2-005) can be configured to emit light at two, three, four, five, or more distinct wavelengths. In certain embodiments, the wavelength(s) of light emitted by the light source are predetermined. For example, but not by way of limitation, the wavelength(s) can be predetermined such that, when irradiated onto a sample, e.g., a sample containing a fluorescent protein, the emitted light induces fluorescence. In certain instances, the wavelength(s) used in connection with the methods described herein will fall within the range of 400 nm to 650 nm. In certain instances, the light source (2-005) emits light having a wavelength of 400 nm to 408 nm, 550 nm to 565 nm, or 638 nm to 650 nm. In certain non-limiting embodiments, the light source (2-005) is configured to include three lasers with nominal center wavelengths of 405 nm, 560 nm, and 640 nm, which may vary within the absorption band of the fluorophore used. In certain instances, the 405 nm wavelength is used to excite Hoechst dye. In certain instances, the dye attached to the HaloTag (e.g., JF 549 A wavelength of 560 nm is used to excite .

[0059] In certain non-limiting embodiments, the light source (2-005) is used to catalyze a photochemical reaction. For example, but not by way of limitation, the wavelength(s) and irradiation intensity can be such that cleavage of a chemical bond occurs. As a further example, but not by way of limitation, the wavelength(s) and irradiation intensity can induce the adoption of a non-radiative dark state (i.e., "photobleaching molecules"). As a further example, but not by way of limitation, the wavelength(s) and irradiation intensity can induce radiative or non-radiative energy transfer between fluorophores within the sample.

[0060] In certain embodiments of the image acquisition system described herein, the light source (2-005) can be configured to deliver a predetermined amount of power to the back focal plane of the objective lens (2-105). For example, and without limitation, the light source (2-005) can deliver more than 10 mW of power at a particular wavelength, such as 405 nm, and / or more than 150 mW of power at another wavelength, such as 640 nm. Additionally or alternatively, if the light source (2-005) includes three lasers emitting at wavelengths of 405 nm, 560 nm, and 640 nm, respectively, the light source (2-005) can be configured to deliver a predetermined amount of power to the back focal plane of the objective lens (2-105). For example, and without limitation, 405 nm can be configured to deliver more than 10 mW, 560 nm can be configured to deliver more than 150 mW, and 640 nm can be configured to deliver more than 50 mW.

[0061] In certain embodiments of the image acquisition system described herein, the light source (2-005) is configured to emit pulsed light. For example, but not by way of limitation, the light source (2-005) can be configured to emit strobe pulsed light. In certain embodiments of the image acquisition system described herein, the light source (2-005) is configured to emit pulsed light synchronized with the start of image acquisition. In certain non-limiting embodiments, the light source (2-005) pulses at specific time intervals depending on the number of frames per second being captured. For example, but not by way of limitation, if 100 frames per second (FPS) are being captured by the detector (2-165), the laser will be on for 9 ms and off for 1 ms. In contrast, in a 200 FPS mode, the laser will be on for 4 ms and off for 1 ms. In certain embodiments of the OLS htSMT workflow, the light source is configured to transition from 90% to 10% power in less than about 0.4 ms. In a particular embodiment of the OLS htSMT workflow, the light source is configured to transition from 90% to 10% power in less than about 0.2 ms.

[0062] The emission of light by the light source (2-005) and the directing of that light to the optical relay (2-010) may be facilitated using a single mode fiber in certain embodiments of the image acquisition systems disclosed herein. Alternatively, multimode fiber may be used in certain embodiments of the image acquisition systems disclosed herein. For example, but not by way of limitation, the multimode fiber may be configured in a predetermined shape for sample illumination.

[0063] In certain embodiments of the image acquisition systems described herein, e.g., with respect to systems configured for high-throughput sample analysis, the light source (2-005) can be configured to exhibit low drift in output. In certain embodiments, such low-drift configurations increase the consistency of sample processing and facilitate high-throughput analysis. For example, and not by way of limitation, such low-drift output configurations maintain output within about 0% to about 15% variation, about 0% to about 10% variation, about 10% variation, about 9% variation, about 8% variation, about 7% variation, about 6% variation, about 5% variation, about 4% variation, about 3% variation, about 2% variation, or about 1% variation.

[0064] In certain instances, such a low-drift output configuration maintains output power within about 0% to about 15% variation, about 0% to about 10% variation, about 10% variation, about 9% variation, about 8% variation, about 7% variation, about 6% variation, about 5% variation, about 4% variation, about 3% variation, about 2% variation, or about 1% variation across ambient (room) temperature variations of, for example, 17°C + / - 5°C. In certain instances, this is achieved by using a temperature sensor and / or a closed-loop heater to stabilize the temperature of the internal light source (e.g., laser engine) and thereby reduce output power drift. For example, but not by way of limitation, the light source may be thermally isolated from ambient temperature variations using an insulated enclosure design. Additionally or alternatively, closed-loop heaters can be strategically placed in specific locations within the system, such as at the fiber coupler, to reduce output power drift. Additionally or alternatively, a water jacket and / or cooling device can be used to reduce heat buildup from the laser head. Additionally, these thermal controls, used individually or in combination, reduce the warm-up time to reach a steady state of operation and maintain a more stable internal operating temperature as the laser is powered off and on.

[0065] 2.1.2. Optical Elements and Sample Illumination Referring to the exemplary image acquisition system of Figure 2A, the system includes a light source (2-005) configured to emit light, which is relayed by one or more optical elements in an optical relay (2-010) configured to shape the light emitted from the light source to form a shaped beam (2-065). The particular optical elements of any particular optical relay (2-010) embodiment can be selected and configured to not only generate a beam (2-065) of an appropriate shape, but also to provide appropriate translation of that beam.

[0066] In certain non-limiting embodiments of the optical relay (2-010) of the image acquisition system of the present disclosure, the optical relay (2-010) comprises one or more lenses and / or other optical elements. For example, but not by way of limitation, the selection and orientation of the lenses and other optical elements in the optical relay (2-010) are configured to appropriately shape the light beam directed at the sample. In certain non-limiting embodiments, the optical relay (2-010) includes an optical element, such as a collimator (2-020), for collimating the light emitted from the light source (2-005). Additionally or alternatively, the optical relay (2-010) may include additional optical elements, such as a Powell lens (2-025) or other element adapted to create a beam fan, one or more cylindrical lenses (2-045 and 2-055), one or more slits (2-050 and 2-095) for adjusting the size of the light sheet, one or more achromatic lenses (2-060 and 2-080), and / or one or more mirrors (2-070, 2-075, and 2-085), one or more of which may be galvo mirrors (2-085) that can translate the light. The specific attributes of the optical elements are predetermined to produce an appropriately shaped light beam. For example, but not by way of limitation, the OLShtSMT system of the present disclosure may achieve a uniform horizontal FOV and a uniform vertical FOV. Such uniformity of horizontal and vertical FOV is in contrast to other strategies that provide non-uniform horizontal and / or non-uniform vertical FOV (see Table 2). [Table 2]

[0067] To achieve a uniform horizontal FOV and a uniform vertical FOV, the optical relay (2-010) of the OLS htSMT system described herein includes an optical element or assembly capable of generating a beam that is elongated along the X-plane and narrow along the Y-plane, where the light beam has a uniform intensity across the long dimension of the linear shape. In a specific, non-limiting embodiment, the optical relay (2-010) of the OLS htSMT system described herein includes a Powell lens (2-025) that shapes the light beam to have a uniform intensity across the long dimension of the linear shape (2-065). The optical relay (2-010) of the OLS htSMT system described herein can include additional or alternative optical elements or assemblies that shape the light beam to have a uniform intensity across the long dimension of the linear shape (2-065). For example, but not by way of limitation, the optical relay (2-010) of the OLS htSMT system described herein can include a diffractive element or assembly configured to shape the light beam to have a uniform intensity across the long dimension of the linear shape.

[0068] In certain non-limiting embodiments of the optical relay (2-010) of the image acquisition system of the present disclosure, the optical relay (2-010) comprises one or more optical elements or assemblies configured to translate the light beam relative to the sample surface of the sample being analyzed, e.g., in a direction perpendicular to the long dimension of the light beam. For example, and without limitation, such optical elements or assemblies configured to translate the light beam relative to the sample surface of the sample being analyzed can include a galvo mirror (2-085) or a piezoelectric element configured to translate the light beam. Additionally or alternatively, such optical elements or assemblies configured to translate the light beam relative to the sample surface of the sample being analyzed can comprise a computer-controlled motor.

[0069] Referring to the exemplary image acquisition system of FIG. 2A, the system includes an optical relay (2-010) configured to shape light emitted from a light source to form a shaped beam (2-065), which is directed by an optical element (2-100), such as a dichroic mirror, and directed towards an objective lens (2-120), thereby illuminating a sample surface (2-130) with an oblique beam (2-125).

[0070] In certain non-limiting embodiments of the image acquisition system of the present disclosure, the objective lens (2-120) directs an inclined beam (2-125) onto the sample surface (2-130) to be analyzed. In certain non-limiting embodiments of the image acquisition system of the present disclosure, the objective lens (2-120) is a water-immersion objective lens. The use of a water-immersion objective lens facilitates high-throughput sample analysis by eliminating the oil present in conjunction with the use of an oil-immersion objective lens, thereby enabling higher quality and less distortion. In the context of automated systems, the presence of oil can be problematic, as well as the potential for oil to spread to components such as optical elements that may be exposed to oil and become contaminated. However, water-immersion objective lenses have a refractive index more suited to cellular imaging, resulting in less distortion and improved image quality than oil-immersion objective lenses. In certain non-limiting embodiments, the objective lens is a 60x 1.27 NA water-immersion objective lens (Nikon). In certain embodiments of the workflow described herein, the water-immersion objective lens (2-120) is heated by a heating element. For example, such a heating element maintains the water immersion objective (2-120) at a temperature sufficient to avoid inducing temperature changes in the sample contained in the sample plate (2-021).

[0071] 2.1.3. Image Acquisition In certain non-limiting embodiments of the image acquisition system of the present disclosure, the objective lens (2-0120) is also used to focus the fluorescence emitted by the sample (2-145) in response to the illumination provided by the oblique beam (2-125). In certain non-limiting embodiments, the fluorescence emission (2-145) focused on the objective lens passes through absorption filters (2-150 and 2-160), e.g., bandpass absorption filters that match the spectrum of the fluorophore under observation and are mounted on a high-speed filter wheel (Finger Lakes Instruments), and is collected by a detector device (2-165). In certain non-limiting embodiments, the fluorescence emission focused on the objective lens is directed to an optical relay before collection by the detector device (2-165). For example, but not by way of limitation, such an optical relay can include one or more lenses (2-155) and one or more additional optical elements, e.g., elements configured to reject additional scattered light before collection by the detector device (2-165). In certain non-limiting embodiments, the fluorescent emission focused onto the objective is directed through another dichroic mirror to split the emission across multiple regions of the detector (2-165). In certain non-limiting embodiments, the fluorescent emission focused onto the objective is directed through another dichroic mirror to split the emission across multiple detectors (2-165).

[0072] In certain non-limiting embodiments of the image acquisition system of the present disclosure, the detector device is configured to synchronize detection of the tilted beam (2-125) across the sample plane (2-130) with the movement of the beam. Such synchronization is shown schematically in FIG. 2F. For example, but not by way of limitation, the detector device can be a CMOS camera, such as a back-illuminated CMOS camera (the Hamamatsu Fusion BT).

[0073] In certain embodiments of the image acquisition system of the present disclosure, the CMOS camera can be operated to collect a series of SMT frames for each field of view. For example, but not by way of limitation, 1-20,000 SMT frames, 1-15,000 SMT frames, 1-10,000 SMT frames, 1-5,000 SMT frames, 1-1,000 SMT frames, 2-500 SMT frames, 5-250 SMT frames, 10-200 SMT frames, 100-200 SMT frames, or 200 SMT frames can be collected per field of view. In certain embodiments, the CMOS camera can be configured to operate at a frame rate of 0.5-1000 Hz, although in certain embodiments, it can be configured to operate at 100 Hz. For example, but not by way of limitation, certain cellular SMT implementations can be performed at 100 Hz.

[0074] In certain non-limiting embodiments of the image acquisition system of the present disclosure, the detector device is configured to transmit a signal at each frame to trigger other elements of the imaging system. For example, but not by way of limitation, the detector device can trigger illumination from the light source (2-005) to collect fluorescent emission associated with a strobe laser pulse. For example, but not by way of limitation, such fluorescent emission collection can be associated with a 10-100 millisecond frame and a 2 millisecond strobe laser pulse.

[0075] In certain embodiments, the imaging acquisition system can be configured to acquire a predetermined image dimension per frame, referred to herein as a region of interest (ROI). In certain embodiments, the ROI varies depending on the frame rate employed. For example, at 100 FPS, 2304 x 1728 pixels define an ROI that is 248.832 x 186.624 microns at the sample plane. In contrast, at 200 FPS, 2304 x 768 pixels define an ROI that is 248.832 x 82.944 microns at the sample plane.

[0076] In certain embodiments, the imaging acquisition system can be configured to perform a predetermined sweep speed at a predetermined frame rate. For example, and not by way of limitation, at 100 FPS, the sweep speed may be 186.624 microns / 9 ms, which corresponds to 20.8 microns / ms, which corresponds to 2.08 cm / s. In contrast, at 200 FPS, the sweep speed may be 82.94 microns / 4 ms, which corresponds to 20.7 microns / ms, which corresponds to 2.07 cm / s.

[0077] In certain embodiments, a detector device can be used to collect fluorescent emissions at multiple wavelengths. For example, but not by way of limitation, fluorescent emissions of additional fluorophores can be collected at the same frame rate or at different frame rates for the same field of view to provide downstream registration of SMT tracks to other cellular components, such as nuclei. Additional channels of the detector device can be used as needed to expand the number of fluorescent emissions simultaneously captured for the same field of view to provide downstream registration of SMT tracks to other cellular components, such as nuclei.

[0078] 2.2. Sample Handling Referring to FIG. 1 , embodiments of the present subject matter can be implemented using an htSMT workflow incorporating a system for sample preparation, including reagent processing. For example, and not by way of limitation, FIG. 3 provides a schematic diagram of a sample plate (2-021) including multiple wells (2-016) in which sample preparation and analysis can occur. FIG. 3 also provides a schematic diagram of the components of a sample, e.g., cells (2-018) and fluorescent target proteins (2-017) within the cells. However, as noted herein, FIG. 3 is not intended to convey scale; for example, each sample present in well (2-016) may contain thousands of cells, each cell containing multiple fluorescent target proteins. FIG. 3 also schematically illustrates the ability of the sample processing system of the present disclosure to add additional reagents to the sample (2-019). The addition of such reagents can be handled by robotic manipulation, including, but not limited to, translation of a robotic fluid handling system relative to the individual wells (2-016) of the sample plate (2-021), translation of the sample plate (2-021) itself, or a combination of both. In certain embodiments of the image acquisition system, the sample plate (2-021) can be maintained in a temperature-controlled environment via the environmental control area (2-020). For example, but not limited to, the samples can be maintained at 22-50°C. In certain embodiments of the image acquisition system, the sample plate (2-021) can be maintained in a humidity-controlled environment via the environmental control area (2-020). For example, but not limited to, the samples can be maintained at 20%-95% humidity. In certain embodiments of the image acquisition system, the sample plate (2-021) can be maintained in a defined gas environment via the environmental control area (2-020). For example, but not limited to, the samples can be maintained at 5% CO2.

[0079] 2.2.1. Cell Lines and Cell Culture Referring to FIG. 3, a particular advantage of the htSMT system described herein is its ability to assay live cells (2-016), facilitating tracking of protein activity, mobility, and diffusion behavior within a dense, live-cell environment. Exemplary cell lines for use in conjunction with the htSMT system described herein are contemplated if the sample can be focused by the objective lens (2-120) for a sufficient period of time to direct the fluorescent emission of the fluorophore toward the detector (2-165). For example, but not by way of limitation, cells can be directly attached to a coverslip. As a further example, but not by way of limitation, cells can be induced to attach to the coverslip after treating the coverslip with an extracellular matrix material (e.g., fibronectin, collagen, poly-D-lysine, laminin, Matrigel, vitronectin, etc.). As a further example, but not by way of limitation, cells can be induced to attach to the coverslip after treating the coverslip with plasma. Exemplary cell lines can be selected to minimize non-fluorophore emissions reaching the detector. For example, and not by way of limitation, particular cell lines used in connection with the htSMT system described herein include U2OS cells (ATCC Catalog No. HTB-96), MCF7 cells (ATCC Catalog No. HTB-22), T47d cells (ATCC Catalog No. HTB-133), and SK-BR-3 cells (ATCC Catalog No. HTB-30).

[0080] In certain embodiments of the htSMT system of the present disclosure, the cells used are cultured as needed to provide sufficient cell numbers to achieve the desired high-throughput analysis. For example, but not limited to, cells such as U2OS cells (ATCC Catalog No. HTB-96), MCF7 cells (ATCC Catalog No. HTB-22), T47d cells (ATCC Catalog No. HTB-133), and SK-BR-3 cells (ATCC Catalog No. HTB-30) can be grown in DMEM (Cat. No. 1056601, Gibco DMEM, High Glucose, GlutaMAX Supplement, ThermoFisher) supplemented with 10% fetal bovine serum (Cat. No. 16000044, ThermoFisher) and 1% penicillin-strep (Cat. No. 15140122, ThermoFisher), maintained in a humidified 37°C incubator with 5% CO2, and subcultured approximately every 2-3 days. Additional culture strategies that may be suitable for use with the cell lines outlined herein will be known to those of skill in the relevant art.

[0081] In certain embodiments of the htSMT system of the present disclosure, cells contain one or more fluorescent target proteins. The choice of the specific protein(s) to be labeled and the specific labeling approach can vary depending on the particularities of a particular investigation. For example, but not by way of limitation, one approach for labeling proteins used in connection with the htSMT system described herein is the HaloTag fusion strategy. For example, but not by way of limitation, one approach for labeling proteins is SNAPtag fusion. For example, but not by way of limitation, one approach for labeling proteins is to use a fluorophore ligase system. For example, but not by way of limitation, one approach for labeling proteins is to use a tetracysteine ​​motif such as FlAsH or ReAsH. For example, but not by way of limitation, one approach for labeling proteins is by strain-promoted alkyne-azide cycloaddition of a fluorophore. For example, but not by way of limitation, one approach for labeling proteins is by inducing cellular uptake of a separately produced fluorescent target protein. In certain embodiments of the htSMT system of the present disclosure, the cells contain one or more fluorescently labeled glycoproteins.

[0082] While those skilled in the art can implement the HaloTag fusion approach in a variety of ways, one exemplary approach is to transfect a mammalian expression vector containing a fusion gene (i.e., a protein of interest fused in frame with the HaloTag sequence) under the control of a weak L30 promoter and containing a neomycin resistance marker into a cell line of interest (e.g., U2OS cells). In certain embodiments, such transfection can be achieved when cells are at 70% confluence using FuGENE6 (Cat. No. E2691, Promega). In certain embodiments, transfected cells can then be selected using an appropriate selection agent, e.g., G418 (Cat. No. 10131027, Thermo Fisher), at an appropriate concentration, e.g., 500 μg / mL. In certain embodiments, cells can then be clonally isolated. Clones expressing the desired fusion gene can be initially transfected with 100 nM JF 549 -HTL (Cat. No. GA1110, Promega) and 50 nM Hoechst 33342 staining, JF 549 This can be determined by identifying clones with the expected distribution of signals. In certain embodiments, 3-6 clones can then be tested for response to a control compound using SMT conditions, and the most homogeneous clones can then be expanded for further testing.

[0083] Although the htSMT workflow of the present application is generally described with respect to an embodiment tracking the effects of compounds on a target fluorescent protein, the htSMT workflow described herein is equally applicable to tracking and analyzing fluorescent target compounds. For example, but not by way of limitation, the compounds described herein may themselves be fluorescent or may be modified to facilitate fluorescent detection. Furthermore, changes in the motion of fluorescent compounds can be used to determine the SMT profile of the compound itself. Thus, all analytical strategies described herein for tracking a target fluorescent protein are also applicable to results obtained by tracking the compound itself.

[0084] 2.2.2. Single Molecule Tracking Sample Preparation Referring to FIG. 3, embodiments of the present subject matter can be implemented using an htSMT workflow whereby cells (2-018) are seeded onto plates (2-021), e.g., tissue culture-treated 384-well glass-bottom plates, although other plate types, including but not limited to single chamber, 9-well glass-bottom plates, 24-well glass-bottom plates, 96-well glass-bottom plates, 1536-well glass-bottom plates, and 3456-well glass-bottom plates, as well as plates made of alternative materials, e.g., plates made partially or entirely of plastic, can also be used in conjunction with the approach outlined herein. In certain embodiments, cells (2-018) are seeded at 1 to 20,000 cells per well (2-016), e.g., 50 to 10,000, 100 to 9,000, 250 to 8,500, 500 to 7,500, 750 to 7,000, 2,500 to 6,500, or 6,000 cells. The seeded cells can then be incubated under conditions favorable for attachment, e.g., overnight at 37°C and 5% CO2. Cells can be incubated with a sufficient amount of label to allow fluorescence, e.g., in the case of HaloTag fusions, 0.1 to 100 pM of JF 549 The desired results can be achieved by incubating -HTL (catalog no. GA1110, Promega) with 50 nM Hoechst 33342 (for labeling nuclei) in complete medium for 1 hour.

[0085] In certain embodiments of the htSMT strategy described herein, the cells are then washed, for example, three times in DPBS and twice in imaging medium. In certain embodiments, the imaging medium is prepared to facilitate fluorescence, for example, fluoroBrite DMEM medium (catalog number A1896701, Thermo Fisher), which can be supplemented with GlutaMAX (catalog number 35050079, Thermo Fisher) and the same serum and antibiotics as the growth medium.

[0086] If appropriate, compounds can be added to samples to test their effect on specific target proteins via SMT. In certain embodiments, compounds can be serially diluted in an Echo-certified 384-well low-dead-volume source microplate (product number 0018544, Beckman Coulter) to generate source material for dose titration. Compounds can then be administered to cell culture media at a final dilution of, for example, 1:1000. In certain embodiments of the htSMT strategies described herein, each dose of compound has at least two replicates per plate and three plate replicates. Additionally, in certain embodiments of the htSMT strategies described herein, 20 DMSO control wells and two no-dye control wells can be randomized across each sample plate (2-020). In certain embodiments, compounds can be incubated for 0-48 hours, for example, 1 hour at 37°C, before acquiring images.

[0087] 3.OLS htSMT software 3.1. htSMT Software Overview FIG. 4 illustrates an exemplary system 400 for a high-throughput single-molecule imaging platform for measuring molecular movement in living cells. An experiment 402 can be performed to collect large amounts of data from multiple living cells (e.g., using an imaging system 424 to identify compounds 426 and / or targets 422). The experiment 402 can include applying various identifiers, such as labels that can later fluoresce or otherwise be detected, to molecules of interest (e.g., using a laser or other light source). Biological samples forming part of such an experiment 402 can be organized in a plate 404 having multiple wells 406. Each well 406 can have one or more associated fields of view (FOVs) 410. An FOV 410 can be a position within or corresponding to a single well 406. Image sequences can be generated for the FOV 410 to generate one or more movies 412, which can include SMT movies and non-SMT movies. SMT movies can be used to track the path of individual labeled molecules, such as proteins, generating multiple trajectories. Each trajectory may consist of multiple spots 414 containing spatiotemporal coordinates of labeled molecules at a particular time (as described in more detail in FIG. 5). Separately from, and in some examples in parallel with, tracking, the video 412 can be used to identify molecules by using machine learning and / or computer vision-based image segmentation to generate masks 418. Masks 418 are spatial regions within the FOV 410 generated by segmentation. Each mask 418 can belong to a mask category, which is described in more detail in FIG. 4.

[0088] Data associated with the two channels (e.g., the tracking channel and the segmentation / masking channel) can be combined to generate multiple metrics 420 associated with various aspects of the sample. In other words, the trajectories 416 (e.g., trajectory data) can be combined with the machine learning processed image segmentation data and further analyzed using statistical / machine learning techniques. Processing of the combined data can be used to generate metrics 420, such as hit scores, associated with compounds and / or targets within the biological sample, which may be stored in a database structure, as further described in FIG. 5.

[0089] FIG. 5 illustrates data flow through an exemplary system 500 for a high-throughput single-molecule imaging platform for measuring protein movement in live cells. An experiment specification 504 defining an experiment 502 can be provided as data input via one or more clients 702. For example, each experiment 502 can be collected along with an accompanying stain (e.g., Hoechst or Potomac Red) used for downstream analysis, including segmentation 418. The experiment specification 504 can define various parameters for the experiment 402, such as stains, dyes, compounds, and treatments. As previously described in FIG. 4, the imaging system 506 (e.g., imaging system 424) can capture a sequence of images to generate one or more SMT movies 511 and / or non-SMT or segmented movies 508 (e.g., movie 412) that characterize the movement of molecules. The SMT movies 511 can characterize the movement of individual fluorophores and / or can include images of individual fluorophores. The segmented movies 508 can include a sequence of images that characterize the movement of labeled molecules and / or their components. It will be understood that Hoechst staining is only one technique that may be used to label molecules, and / or multiple labeling techniques, such as Potomoc Red, may be utilized depending on the desired configuration. For example, MitoTracker Deep Red may be used to label mitochondria, Concanavalin A-dye conjugates may be used to label endoplasmic reticulum, SYTO14 may be used to label nucleoli, phalloidin may be used to label actin, etc.

[0090] The SMT movie 511 can be analyzed to perform operations related to molecule tracking 510, which can include detection 512, sub-pixel localization 513, and linking 514 to identify trajectories 515 of molecules across various images in the SMT movie 511. More specifically, during detection 512, one or more spots can be detected or recovered in the SMT movie 511. Each spot can be provided with spatiotemporal coordinates. These spatiotemporal coordinates can be estimated by using sub-pixel localization techniques 513. Linking 514 can be performed on the spots to ultimately identify the trajectories 515.

[0091] As used herein, a link is a potential connection between two spots. Each link is directed, starting at one spot and ending at another. A "correct link" connects two spots generated by the same emitter in different frames; otherwise, the link is "incorrect." One goal of the linking algorithm is to estimate which link is correct. In this specification, links are referred to in the form a:i→j, which is interpreted to mean link a starting at spot i and ending at spot j. A link satisfies at least three of the following constraints: (a) the link moves forward in time; (b) the link cannot connect two spots that are more distant than a certain limit (hereinafter referred to as the "search radius"); and (c) the link cannot connect two spots that are more distant in time than a certain limit (hereinafter referred to as the "gap limit"). A spot-link graph is a graph of the spots and links of a single SMT video 511. In this graph, spots are vertices and links are edges. Because links move forward in time, the spot-link graph is a directed acyclic graph. A matching is a subset of links in the spot-link graph, such that no two links in this subset start or end at the same spot. A trajectory 515 is used herein to refer to a sequence of consecutive (end-to-end) links in the same matching. A dynamic metric 530 can be determined using multiple trajectories. Such parameters can include spot attributes that characterize the spot's movement. Such parameters can include one or more of the velocity, diffusion coefficient, or anomaly parameter(s) of each spot. The dynamic parameter(s) of spot i are defined herein as θ i Herein, the set of dynamic parameters of all spots in the spot-link graph is called Θ.

[0092] Separately from, and in some variations in parallel with, the processing of the SMT video 511, the segmented video 508 can undergo segmentation to generate one or more masks 520. The masks can be classified into various categories, including, but not limited to, cell nuclei, cytoplasm, and / or extraneous masks, which are further described in FIG. 6 . An instance mask is an individually segmented object (e.g., one cell, one nucleus, one mitochondrion). The FOV 410 can include any number of instance masks for one mask category. A semantic mask is the union of all instance masks corresponding to one type of mask category for one FOV (e.g., all cells, all nuclei, or all mitochondria for one FOV). Extraneous masks can include portions of the non-SMT video 508 that are excluded from any downstream data analysis. For example, these extraneous masks can correspond to portions of the non-SMT video 508 that are out of focus or contain autofluorescent cellular debris that prevents accurate tracking. During segmentation, molecules in the segmented movie 508 can be assigned to one or more masks. Image metrics 540 can be assessed from the masked molecules, such as cell health, focus quality, etc.

[0093] Experiment information, such as dynamic metrics 530, image metrics 540, and any data from which any metrics are derived (e.g., segmentation information), may be provided to a data repository 570 for storage. Such a data repository 570 may store any results of the experiment 402, such as, for example, dynamic metrics 530, image metrics 540, and / or any data from which any metrics are derived. The data repository may include a local persistence server and / or a dedicated server accessed locally or via the cloud. The data repository 570 may also store metadata associated therewith and / or associated with the experiment specification 504. Experiment information (e.g., results and metadata of past experiments, etc.) may be provided to the data repository 570 via a repository application program interface (API) 550. The repository API 550 may also interface with a web-based graphical user interface front end 560 that provides such information for display on the client 502.

[0094] In some variations, the segmentation information can be used to identify subcellular compartments such as the nucleus, nucleolus, cytoplasm, etc. The segmentation information may also be used to distinguish one cell from another. The segmentation information may be stored in a particular format (e.g., a multi-image file format such as TIFF, etc.).

[0095] The exemplary dynamics metric 530 can also include a state array. The state array is a framework for learning interpretable dynamic models from SMT trajectories and can be used to gain additional insight into the movement of a target protein and where that movement occurs within the cell. In some variations, the state array can be generated / added using segmentation information. The state array output can be returned at the subcellular compartment level, allowing researchers to distinguish between dynamics in different subcellular compartments. Additionally, the state array can be calculated for each individual subcellular compartment (e.g., per nucleus).

[0096] To facilitate data access by applications, including but not limited to state arrays, processed SMT data may be stored in a format that allows for (a) representation of processed trajectories and associated attributes, such as the SNR and spot shape characteristics of each SMT video; (b) representation of mask objects, including mask categories (e.g., the subcellular organelles associated with each mask object); (c) association of trajectories with mask objects (e.g., the cell nuclei in which each trajectory was observed); and (d) association of all SMT videos with metadata related to the original experiment, such as compound treatment, acquisition time, and imaging system name. Formats (a) and (c) may be protocol buffer schemas that define the storage format for trajectories and associated mask objects. Format (b) may be a specialized image file format containing the mask object to which each pixel in the FOV belongs. Format (d) may be a PostgreSQL database that records all captured experiments / videos. As a client of the processed SMT data, state arrays can reference these data schemas to report the dynamic properties of trajectories by mask category or by mask object.

[0097] FIG. 6 illustrates multiple images 600 showing the differences between mask categories and instance or semantic masks. As previously discussed, a non-SMT video or segmented video can be assigned to multiple categories. Such categories may include cell nuclei (e.g., Category A), cytoplasm (e.g., Category B), and / or extraneous masks (e.g., Category C). Unique, individual masks can be applied to a biological sample. For example, image 610 illustrates a unique, individual instance mask applied to a cell nucleus (e.g., Category A). Image 612 illustrates a unique, individual instance mask applied to a cell cytoplasm (e.g., Category B). Image 620 illustrates multiple instance masks applied to one or more nuclei, each with a different color representing a unique, individual instance mask. Image 622 illustrates multiple masks applied to one or more cytoplasms, each with a different color representing a unique, individual instance mask. Image 630 illustrates a semantic mask that is the union of all instance masks applied to one or more nuclei. Image 632 illustrates a semantic mask applied to one or more cytoplasms.

[0098] FIG. 7 illustrates an exemplary computer-implemented environment 700 in which an imaging system 710 can interact with a computing architecture to execute various algorithms described herein. As shown in FIG. 7 , the imaging system 710 can interface with one or more clients 750 (e.g., client 502 via a web application having a graphical user interface). The one or more clients 750 can interface with one or more servers 720 accessible via network(s) 730. The one or more clients 750 can host frame grabbers that capture images (e.g., video 412) from a camera. These images can be temporarily stored on the one or more clients 750 and periodically transferred via the network 730 to the one or more servers 720 for remote storage. The one or more servers 720 also include or have access to one or more data stores 740 for storing data collected and / or extracted from the sample by the imaging system 710. In some variations, the network 730 may include or interface with one or more network storage arrays 760 for storing data such as captured images (e.g., video 412).

[0099] FIG. 8 is a diagram 800 illustrating a sample computing device architecture for implementing various aspects described herein. In some variations, the sample computing device architecture may be that of a client(s) 750 and / or a server(s) 720, and some components described in connection with diagram 800 may be optional for the client(s) 750 and / or the server(s) 720. A bus 804 may serve as an information highway interconnecting the other illustrated components of hardware. A processing system 808 (e.g., one or more computer processors / data processors in a given computer or in multiple computers), labeled CPU (Central Processing Unit), may perform the computational and logical operations required to execute a program. Optionally, or additionally, a processing system 812 (e.g., one or more computer processors / data processors in a given computer or in multiple computers), labeled GPU (Graphics Processing Unit), may perform the computational and logical operations required to execute a program. Non-transitory processor-readable storage media, such as read-only memory (ROM) 816 and random access memory (RAM) 820, may be in communication with processing system 808 and / or processing system 812 and may contain one or more programming instructions for the operations specified herein. Optionally, the program instructions may be stored on a non-transitory computer-readable storage medium, such as a magnetic disk, optical disk, recordable memory device, flash memory, solid-state drive, or other physical storage medium.

[0100] In one example, disk controller 848 can interface with one or more optional removable storage 856 or local storage 852 via system bus 804. Removable storage 856 can be an external or internal disk drive, or a solid-state drive, or an external hard drive. Local storage 852 can be an internal hard drive and / or memory. As mentioned above, these various examples of removable storage 856, local storage 852, and disk controller 848 are optional devices. System bus 804 may also include at least one communication interface 824 to enable communication with external devices either physically connected to the computing system or externally available via a wired or wireless network, such as cloud storage or a remote service. In some cases, at least one communication interface 824 includes or otherwise comprises a network interface.

[0101] In some variations, e.g., client(s) 750, to provide for interaction with a user, the subject matter described herein can be implemented on a computing device having a display device 844 (e.g., an LCD (liquid crystal display) or LED (light emitting diode) monitor, etc.) for displaying information retrieved from bus 804 to a user via a display interface 840, and an input device 832, such as a keyboard and / or pointing device (e.g., a mouse or trackball) and / or a touch screen, for a user to provide input to the computer. Other types of input device 832 can also be used to provide for interaction with a user. For example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback via microphone 836, or tactile feedback), and input from the user can be received in any form, including acoustic, voice, or tactile input. Input device 832 and microphone 836 can be connected to bus 804 via input device interface 828 to communicate information. By way of example, input device 832 can be an imaging system 710 configured with the capability to capture a sequence of images, as described herein. Frame grabber 858 can capture or grab individual frames from analog or digital data encapsulating a sequence of images acquired from bus 804. Frame grabber 858 can include memory capable of storing single or multiple frames. Frame grabber 858 can also provide individual frames or multiple frames to bus 804 for further storage, for example, in local storage 852 and / or removable storage 856. Other computing devices, such as dedicated servers, can omit one or more of the components described in connection with FIG. 8.

[0102] One or more aspects or features of the subject matter described herein may be implemented in digital electronic circuitry, integrated circuits, specially designed application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), computer hardware, firmware, software, and / or combinations thereof. These various aspects or features may include implementation in one or more computer programs executable and / or interpretable by a programmable system including at least one programmable processor, which may be special-purpose or general-purpose, coupled to receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device. The programmable system or computing system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and by virtue of the client-server relationship they have to each other.

[0103] These computer programs, which may also be referred to as programs, software, software applications, applications, components, or code, contain machine instructions for a programmable processor and may be implemented in high-level procedural languages, object-oriented programming languages, functional programming languages, logic programming languages, and / or assembly / machine languages. As used herein, a "machine-readable medium" refers to any computer program product, apparatus, and / or device (e.g., magnetic disk, optical disk, memory, programmable logic device (PLD), etc.) used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives the machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor. A machine-readable medium may non-transitory store such machine instructions, such as, for example, a non-transitory solid-state memory, a magnetic hard drive, or any equivalent storage medium. Alternatively or additionally, a machine-readable medium may temporarily store such machine instructions, such as, for example, a processor cache or other random access memory associated with one or more physical processor cores.

[0104] 3.2. Pixel-by-pixel cell health scoring using machine learning Cell health and specific health-related subtypes can be scored or otherwise characterized using machine learning on a pixel-by-pixel basis, which can then be used to characterize various events throughout the image sequence generated by the microscope system. This information can be used to characterize aspects of the cell's life cycle. Using the present subject matter, cellular perturbations with known outcomes can be imaged, and the resulting images can be labeled with those outcomes. In some variations, one or more machine learning models can be trained using the image-level descriptors / labels, allowing these machine learning models to later classify individual pixels in the images. This type of training, described further below, can be characterized as weakly supervised.

[0105] In some examples, machine learning models can be trained and deployed to identify cell health as a function of compound treatment or other environmental factors in Hoechst-labeled fluorescent microscopy images. Such a configuration is particularly advantageous in that the microscopy systems described herein can generate large amounts (e.g., terabytes) of data per day, making analysis computationally expensive and time-consuming.

[0106] A training set can be generated by exposing cells (e.g., U2OS cells, HCT116, etc.) to various concentrations of a first compound and a second compound (e.g., staurosporine, sorbitol, etc.) for various compound incubation times. Figure 9 is a diagram 900 showing healthy cells 910 and unhealthy cells 920. In some variations, the techniques provided herein can be used to characterize protein dynamics and / or protein expression, which can be used to identify whether a particular cell is healthy or unhealthy (or some relative measure).

[0107] FIG. 10 is a diagram 1000 showing a first image 1010 (corresponding to the FOV) of healthy cells, a second image 1020 of fewer healthy cells after the sample has been exposed to sorbitol, and a third image 1030 of fewer healthy cells, similarly after the sample has been exposed to staurosporine.

[0108] To generate a training dataset, microscopic images with sufficient resolution can be generated from a sample of cells as the sample is exposed to various compounds and / or various environmental factors for different periods of time, which can then be used to train one or more machine learning models (e.g., using weak supervision).

[0109] 11 includes a diagram 1100 illustrating a multi-layer machine learning model 1120 (e.g., a U-Net architecture) into which a microscopic image 1110 is input. The layers (e.g., input layer, hidden layer, and output layer) of the machine learning model 1120 include various nodes / neurons with various weights that are generated through a training process. Each pixel in the microscopic image 1110 includes a value that characterizes color and intensity, and the output of the machine learning model 1130 includes a classification or other characterization of the cell health of each pixel (or alternatively, that the pixel is not associated with a cell).

[0110] In one embodiment, U2OS and HCT116 cells were stained with Hoechst 33342 and treated with staurosporine at concentrations ranging from 5 pM to 500 nM and sorbitol at concentrations ranging from 0.1 to 2 M. The cells were then incubated at 37°C and 5% CO2 for 3 hours to induce varying degrees of cell damage. The cells were then imaged using the super-resolution microscope platform used for SMT acquisition, and Hoechst frames were retained for training. Images of nuclei treated with high concentrations of staurosporine or sorbitol were labeled as unhealthy (e.g., second and third images 1020, 1030), while DMSO-treated nuclei were labeled as healthy (e.g., first image 1010). These FOV-level labels were used to weakly supervise pixel-level machine learning models (e.g., machine learning model 1120). Various types of machine learning models can be used in this context, including neural networks such as convolutional neural networks, including U-Net architectures. The machine learning model 1120 may be optimized using a loss function such as weighted categorical cross-entropy or other loss functions that may be used to address imbalances across categories / classes (e.g., healthy / unhealthy / background pixels). The machine learning model 1120 may work in conjunction with a separate segmentation model to label all cells using summary statistics such as mean / median pixel scores per mask instance.

[0111] The training dataset can, in some variations, be augmented with additional images, such as images with different brightness levels. As an example, to ensure that the machine learning model 1120 is robust to varying brightness levels between images, a logarithmic scale can be applied to all training images, along with a pixel intensity multiplier of 70% to 130% across approximately 50% of the training images. Example images 1210-1230 with varying brightness levels are shown in diagram 1200 of FIG. 12.

[0112] Figure 13 is a diagram 1300 illustrating the results of weakly supervised learning, including input images 1310, 1330, 1350 and corresponding outputs 1320, 1340, 1360. The outputs 1320, 1340, 1360 visualize the pixel-level scores produced by the model, shown on a continuous color scale describing the confidence that the pixel belongs to a healthy or unhealthy nucleus.

[0113] Referring to images 1410-1460 in diagram 1400 of Figure 14, to further improve performance, a subset of images can be manually labeled at the pixel level with three classifications: healthy cell, unhealthy cell, and background. The final pixel-level score can correspond to whether the pixel is likely to be a healthy cell, an unhealthy cell, or part of the background. Precision, in this context, refers to the proportion of correct positive identifications, and recall, in this context, refers to the proportion of positives that are actually correctly identified.

[0114] The machine learning model 1120 was validated by testing its ability to score FOVs for select cell death. Three different engineered cell lines seeded at a 1:1:1 ratio for SMT imaging were used. Two cell lines were engineered to be G418 resistant, and the third cell line was not. Cells were seeded in media containing G418 at various concentrations. Individual cell lines were also seeded on separate plates. The cell health model assumed that the reduction in cell health of the mixed cell population would be measured as one-third of that measured in plates of individual cell lines that were not G418 resistant. Figure 15 is a diagram 1500 showing cell health from the mixed population 1510 (i.e., engineered cell lines with G418 resistance) and the cell health of individual cell lines 1520 that were not G418 resistant.

[0115] Various experiments confirm that the machine learning modeling approach to cell health characterization provided herein allows for a deeper understanding of changes in kinetics measured by SMT, including the detection of small effects on cell health that may distort protein dynamics in a nonbiological, pathway-dependent manner. Referring to graph 1600 in Figure 16, first graph 1610 shows an increase in protein dynamics, as measured by mean posterior diffusion coefficient (MPDC), upon treatment with a topoisomerase inhibitor. Upon further investigation, using current technology, referring to second graph 1620, it was found that as compound dosage increased, overall cell health worsened. In other words, graphs 1610 and 1620 show that the increase in MPDC seen with topoisomerase inhibitor concentrations starting at 0.1 is due in part (and perhaps primarily) to an overall decline in cell health. This is an example of a compound that has widespread effects on cell health and is unlikely to directly inhibit the target of interest. Third graph 1630 provides a correlation between first graph 1610 and second graph 1620. Note that the effect is small and not obvious when looking at the untreated images. See diagram 1700 in Figure 17, which includes image 1710 of the DMSO control, image 1720 of 0.123 uM topoisomerase inhibitor, and image 1730 of 10 uM topoisomerase inhibitor.

[0116] In some variations, a subtype machine learning model (i.e., a U-Net architecture, etc.) may be utilized that has a structure similar to machine learning model 1120 of Figure 11. This subtype machine learning model may be part of an ensemble of models that includes machine learning model 1120 of Figure 11, or may be used in a standalone manner.

[0117] In subtype machine learning models, the caspase-3 marker can be used to supervise training. Unlike the general cell health approach described above, which can use weakly supervised learning methods based on compound treatment, methods for detecting cell health subtypes can be fully supervised. Specifically, the caspase-3 marker provides per-pixel intensity that correlates with apoptotic progression.

[0118] A subtype machine learning model can be trained using a dataset prepared by: Identifying nuclear locations from Hoechst images using segmentation methods; Establishing an apoptosis scale of 0 to 1 based on the pixel intensity range from the caspase-3 signal, and Remapping pixel intensities to the apoptotic scale.

[0119] The network parameters of a subtype machine learning model can be optimized using a multi-loss strategy where two independent losses are computed and averaged to create a global loss. Each loss can be based on the class equilibrium cross-entropy, defined as:

number

[0120] The first loss calculates a continuous representation of apoptotic classes, where a value of 0 indicates a cell that has not undergone apoptosis and a value of 1 indicates a cell that has undergone the full progression of apoptosis and died.

[0121] The second loss can calculate a categorical version of the cross-entropy of two classes, i.e., background and nuclei. This means that a model (i.e., a network) is trained to simultaneously segment nuclei and provide a continuous label for apoptosis. Similar techniques can be used to characterize or otherwise train models that score other subtypes, such as oxidative stress and / or inflammation, of corresponding cells.

[0122] FIG. 18 is a diagram 1800 illustrating a first input image 1810, a first general cell health image 1820 corresponding to the output of the cell health machine learning model described above, and a first apoptosis image 1830 corresponding to the output of the subtype machine learning model described above. These images 1810-1830 show a general agreement between the general cell health score and the apoptosis score, indicating that cells within the field of view (i.e., the FOV captured within the first input image 1810) are undergoing apoptosis. FIG. 18 also illustrates a second input image 1840, a second general cell health image 1850 corresponding to the output of the cell health machine learning model described above, and a second apoptosis image 1860 corresponding to the output of the subtype machine learning model described above. In these images 1840-1860, the general cell health score indicates that most cells within the field of view (i.e., the FOV captured within the second input image 1840) are unhealthy. However, most cells are not undergoing apoptosis, and other factors may be stressing the cells.

[0123] FIG. 19 illustrates a process flow diagram 1900 in which, at 1910, a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received, with each image comprising an array of pixels. Each image is then segmented to identify one or more cells at 1920. At 1930, a cell health score is assigned to each pixel of the identified cells based on the corresponding images using at least one machine learning model. At 1940, a total cell health score is calculated based on the cell health scores assigned to the pixels. Data characterizing the calculated total cell health score can then be provided to a consuming application or process at 1950. Related provisioning may include displaying the score, loading the score into memory of a computing device, storing the score in a physically persistent state, and / or transmitting the score over a network to a remote computing system.

[0124] Cell health subtypes can be characterized as part of the process of FIG. 19 or as a separate, standalone process. FIG. 20 illustrates a process flow diagram 2000 in which, at 2010, a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received, with each image comprising an array of pixels. Then, at 2020, each image is segmented to identify one or more cells. At 2030, a cell health subtype score is assigned to each pixel of the identified cells based on the corresponding images using at least one machine learning model. At 2040, a total cell health subtype score is calculated based on the cell health subtype scores assigned to the pixels. Then, at 2050, data characterizing the calculated total cell health subtype score can be provided to a consuming application or process. Related provisioning may include displaying the score, loading the score into a memory of a computing device, storing the score in a physically persistent state, and / or transmitting the score to a remote computing system over a network. The subtypes can characterize one or more of apoptosis, oxidative stress, and / or inflammation.

[0125] 21 is a process flow diagram 2100 in which a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells is received at 2110. Each image is segmented to identify one or more cells at 2120. A cell health score and one or more cell health subtype scores are determined for each identified cell at 2130 using an ensemble of separately trained machine learning models. Data characterizing the scores is then provided to a consuming application or process at 2140. Provided in this regard may include displaying the scores, loading the scores into memory of a computing device, storing the scores in a physically persistent state, and / or transmitting the scores over a network to a remote computing system.

[0126] Microscopic image sequences can visualize fluorescently labeled cellular components within live cells during exposure to compounds or various environmental conditions. Microscopic images can be at the field level. In such variations, the percentage of healthy cells within a field can be calculated (in addition to or as an alternative metric for individual cells).

[0127] Live cells can be stained using a variety of labeling dyes, including Hoechst dyes.

[0128] At least one machine model can be trained using a training dataset. Such a training dataset, in some embodiments, can be generated by selectively exposing a sample of live cells to at least one compound at various concentrations. The compound can take various forms, including, but not limited to, staurosporine and / or sorbitol. The training dataset can also be augmented or otherwise enhanced by randomly varying the brightness levels of the images generated for the training dataset to reflect various environmental conditions and / or by including manually labeled images (either at the field level or pixel level). In some variations, this manual annotation can characterize each pixel as corresponding to one of a healthy cell, an unhealthy cell, or background.

[0129] In certain embodiments, the training data set comprises staining a sample of live cells with a fluorescent dye, exposing a first subset of the sample of live cells to a concentration of a first compound at a first concentration corresponding to unhealthy cells, where the first compound affects cell health, exposing a second subset of the sample of live cells to a concentration of the first compound at a second concentration corresponding to healthy cells, and exposing a third subset of the sample of live cells to a concentration of a second compound at a third concentration corresponding to unhealthy cells, where the second compound affects cell health, and exposing a fourth subset of the live cell samples to a concentration of a second compound at a fourth concentration corresponding to healthy cells; generating images for each of the first, second, third, and fourth subsets of the cell samples using a microscope; labeling the images of the first and third subsets of the live cell samples as corresponding to unhealthy cells; and labeling the images of the second and fourth subsets of the live cell samples as corresponding to healthy cells. The training dataset can be enhanced through other actions, such as varying the exposure time of the first and second subsets of the live cell samples to the first compound and / or varying the exposure time of the third and fourth subsets of the live cell samples to the second compound.

[0130] The machine learning model(s) utilized herein can take a variety of forms, including, but not limited to, pixel-level convolutional neural networks such as those provided by the U-Net architecture. The machine learning model(s) can be optimized using a variety of loss functions, including weighted categorical cross-entropy.

[0131] In certain embodiments, a sequence of microscopic images is received that visualizes fluorescently labeled cellular components within live cells and includes an array of pixels, each corresponding to a specific cell. Each image is segmented to identify one or more cells. Each pixel is then classified as corresponding to a healthy cell, an unhealthy cell, or background using at least one machine learning model. For each identified image, a total cell health score can be calculated based on the classification of the corresponding pixels. Such a total cell health score can be provided to a consuming application or process.

[0132] In certain embodiments, a sequence of microscopic images is received that visualizes fluorescently labeled cellular components within live cells and includes an array of pixels, each corresponding to a specific cell. Each image is segmented to identify one or more cells. Each pixel is then classified as corresponding to a healthy or unhealthy cell using at least one machine learning model trained using weak supervision, in which training images are labeled at the field level (as opposed to the pixel level). For each identified image, an overall cell health score can be calculated based on the classification of the corresponding pixels. Such an overall cell health score can be provided to a consuming application or process.

[0133] 4. Specific OLS htSMT Applications Many, perhaps most, pathways that regulate fundamental cellular biochemistry rely on the interaction of protein sensors and protein effectors that transiently engage and trigger changes in cellular physiology. While the fundamentals of this process have long been recognized, biochemical investigation of these protein interactions has typically required in vitro reconstitution or been investigated through pull-down assays after cell permeabilization. The htSMT workflow described herein provides a means to visualize protein movement in large numbers of live cells, and also in contexts where the effects of additive compounds, such as small molecule inhibitors, can be quantitatively assessed.

[0134] 1 , aspects of the OLS htSMT workflow of the present disclosure include, but are not limited to, (i) sample preparation, including reagent handling, (ii) image acquisition, imaging the sample to generate a series of images and / or video, (iii) image analysis, processing these images and videos, (iv) information storage, and (v) providing insights using the stored information, including biological interpretation. With respect to biological interpretation, the htSMT workflows described herein provide the ability to provide specific insights, as outlined below, depending on the particular workflow employed, e.g., (i) OLS htSMT screening, (ii) OLS htSMT binding, and / or (iii) OLS kinetic SMT.

[0135] 4.1.OLS htSMT Screening In certain embodiments of the OLS htSMT workflow described herein, the systems and methods are adapted to examine the ability of one or more compositions (e.g., "test" compounds) to affect the SMT profile associated with a labeled protein. For example, such htSMT workflows screen for changes in the SMT profile, e.g., either an increase or decrease in the movement of the protein of interest, in the presence of a composition compared to the SMT profile in the absence of the composition. It will also be appreciated that higher-order comparisons can be performed when compounds are multiplexed, including when multiple proteins are fluorescent. Furthermore, as outlined above, the htSMT screening strategies described herein are equally applicable to screening SMT profiles associated with fluorescent compounds, e.g., compounds that are naturally fluorescent or compounds that have been modified to fluoresce or linked to a fluorophore.

[0136] Fundamental to such htSMT screening strategies is the ability of the htSMT workflow described herein to extract large-scale, accurate molecular trajectories. Exemplary OLS htSMT workflows include the following individual strategies and combinations of the following strategies, combining two or more strategic requirements. For example, without limitation, the disclosed workflow includes both illuminating a field of view in a sample plane disposed within a sample with a light beam to induce fluorescence by a subset of fluorescent target proteins in live cells, where the subset of fluorescent target proteins is present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, and illuminating a field of view in a sample plane disposed within the sample with a light beam to induce fluorescence by a subset of fluorescent target proteins in live cells, where the subset of fluorescent target proteins includes proteins in the range of about 1,000 to about 1,000,000. Similarly, illuminating the sample plane to illuminate about 30 to about 80 live cells per FOV and / or generating fluorescence from about 1,000 to about 1,000,000 proteins can be combined with any of the other strategic requirements disclosed herein, such as determining that the average change in fluorescent target protein motion in the presence of a compound is about 5% to about 10% compared to baseline motion under DMSO conditions, detecting fluorescence from multiple fluorescent target proteins within the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system, and achieving a Z-factor greater than 0.5 based on a single field of view.

[0137] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include identifying biological interactions between a compound and a fluorescent target protein in live cells, the workflow including (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein, the workflow further including (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample, the tracking including (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by a subset of the target fluorescent proteins in the live cells, the subset of fluorescent target proteins being dependent on the particular cell type being used. Depending on the target protein, there are approximately 30 to approximately 80 live cells illuminated within the field of view of the sample plane. For example, for U2OS cells, the range is approximately 30 to approximately 40 cells per FOV, while for HCT116 cells, the range is approximately 50 to approximately 80 cells, taking into account differences in area. The tracking further includes (ii) detecting fluorescence from multiple fluorescent target proteins within the field of view of the sample plane via a detector device. The workflow further includes (c) determining changes in the movement of the fluorescent target proteins in the presence of a compound. The change in the movement of the fluorescent target proteins in the presence of the compound compared to the movement of the fluorescent target proteins in the absence of the compound identifies a biological interaction between the compound and the fluorescent target proteins.

[0138] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include identifying biological interactions between a compound and a fluorescent target protein in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample, the tracking including: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by a subset of the target fluorescent proteins in the live cells, the subset of fluorescent target proteins including a range of about 1,000 to about 1,000,000 proteins. The number of proteins in the subset depends on the expression level of the protein of interest and the dye concentration deemed appropriate for labeling the protein subset for robust SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application; the tracking further includes (ii) detecting fluorescence from multiple fluorescent target proteins within a field of view of the sample plane via a detector device; and the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound, wherein a change in the movement of the fluorescent target proteins in the presence of the compound compared to the movement of the fluorescent target proteins in the absence of the compound identifies a biological interaction between the compound and the fluorescent target proteins.

[0139] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include identifying a biological interaction between a compound and a fluorescent target protein in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample, the tracking including: (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to cause fluorescence by a subset of the target fluorescent proteins in the live cells; and (ii) detecting fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane via a detector device; and (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein the average change in the movement of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline movement under DMSO conditions; and a change in the movement of the fluorescent target protein in the presence of the compound compared to the movement of the fluorescent target protein in the absence of the compound identifies a biological interaction between the compound and the fluorescent target protein.

[0140] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include identifying a biological interaction between a compound and a fluorescent target protein in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample, the tracking including: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to cause fluorescence by a subset of the target fluorescent proteins in the live cells; and (ii) detecting via a detector device the fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane; and (iii) tracking including detecting the fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane at a rate of about 10,000 to about 18,000 per day per system; and (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, the change in the movement of the fluorescent target protein in the presence of the compound compared to the movement of the fluorescent target protein in the absence of the compound, thereby identifying a biological interaction between the compound and the fluorescent target protein.

[0141] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include identifying a biological interaction between a compound and a fluorescent target protein in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample, the tracking including: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to cause fluorescence by a subset of the target fluorescent proteins in the live cells; (ii) detecting via a detector device fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane; and (iii) achieving a Z-factor of greater than 0.5 based on the single field of view; and (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein a change in the movement of the fluorescent target protein in the presence of the compound compared to the movement of the fluorescent target protein in the absence of the compound identifies a biological interaction between the compound and the fluorescent target protein.

[0142] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include determining a dose response of a compound to induce a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; and (b) tracking the movement of individual target fluorescent proteins in the plurality of live cells of the sample, the tracking including: (i) illuminating with a light beam a field of view in a sample plane located within the sample to induce fluorescence by at least a subset of the fluorescent target proteins in the live cells; The subset of target proteins is present in approximately 30 to approximately 80 live cells illuminated within the field of view of the sample plane, depending on the specific cell type used. For example, for U2OS cells, the range is approximately 30 to approximately 40 per FOV, while for HCT116 cells, the range is approximately 50 to approximately 80, taking into account differences in area. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device. The workflow further includes (c) determining a change in the movement of the fluorescent target protein in the presence of a compound; and (d) repeating steps (b) to (c) for each of multiple samples across a range of compound concentrations, wherein the change in the movement of the fluorescent target protein across a range of concentrations in the presence of the compound indicates a dose response of the compound.

[0143] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include determining a dose response of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further including: (b) tracking the movement of individual target fluorescent proteins in the plurality of live cells of the sample, the tracking including: (i) illuminating with a light beam a field of view in a sample plane located within the sample to induce fluorescence by at least a subset of the fluorescent target proteins in the live cells; The subset may include proteins ranging from about 1,000 to about 1,000,000, where the number of proteins in the subset depends on the expression level of the protein of interest and the dye concentration deemed appropriate for labeling the subset proteins for robust SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, where the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, where the change in the movement of the fluorescent target proteins in the presence of the compound across the concentration range indicates a dose response of the compound.

[0144] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include determining a dose response of a compound to induce a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further including: (b) tracking the movement of individual target fluorescent proteins in the plurality of live cells of the sample, the tracking including: (i) determining whether the compound induces fluorescence by at least a subset of the fluorescent target proteins in the live cells; (ii) illuminating with a light beam a field of view in a sample plane disposed within the sample; and (ii) detecting fluorescence from one or more fluorescent target proteins in the sample plane via a detector device; the workflow further includes (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein the average change in the movement of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline movement under DMSO conditions; and (d) repeating steps (b) to (c) for each of a plurality of samples across a range of compound concentrations, wherein the change in the movement of the fluorescent target protein across the concentration range in the presence of the compound indicates a dose response of the compound.

[0145] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include determining a dose response of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further including: (b) tracking the movement of individual target fluorescent proteins in the plurality of live cells of the sample, the tracking including: (i) contacting a plurality of live cells of the sample with a compound so as to cause fluorescence by at least a subset of the fluorescent target proteins in the cells; (ii) illuminating with a light beam a field of view in a sample plane disposed within the system; and (ii) detecting via a detector device fluorescence from one or more of the fluorescent target proteins in the sample plane; (iii) said tracking includes detecting fluorescence from a plurality of fluorescent target proteins within the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system; the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, wherein the change in the movement of the fluorescent target proteins in the presence of the compound across the concentration range indicates a dose response of the compound.

[0146] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include determining a dose response of a compound to induce a change in the movement of a fluorescent target protein in a living cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of living cells; (ii) the living cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; and the workflow further including: (b) tracking the movement of individual target fluorescent proteins in a plurality of living cells of the samples, the tracking including: (i) determining a dose response of at least one fluorescent target protein in the cells; (ii) illuminating a field of view in a sample plane located within the sample with a light beam to cause fluorescence by a subset of the fluorescent target proteins; (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device; and (iii) achieving a Z-factor of greater than 0.5 based on the single field of view, the workflow further including (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, wherein the change in the movement of the fluorescent target proteins across the range of concentrations in the presence of the compound indicates a dose response of the compound.

[0147] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include the use of a microscope system configured to identify biological interactions between compounds and fluorescent target proteins in live cells, the system including: (a) a stage for supporting a sample, the sample including a population of live cells, the live cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view in the sample plane, the subset of fluorescent target proteins being present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, depending on the particular cell type used. For example, in the case of U2OS cells, the range is about 30 to about 40 per FOV, while in the case of HCT116 cells, the range is about 50 to about 80, taking into account the difference in area. The system further includes (d) a detector device that monitors a light-based response from the fluorescent target protein in the presence of a compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is located, thereby tracking the position of the fluorescent target protein, and (ii) track the movement of each fluorescent target protein. The system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, the processor being capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0148] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include use of a microscope system configured to identify biological interactions between compounds and fluorescent target proteins in live cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view in the plane of the sample, the subset of fluorescent target proteins including a range of about 1,000 to about 1,000,000 proteins, the number of proteins in the subset being determined based on the expression level of the protein of interest and ... and a dye concentration deemed appropriate for labeling the subset proteins for optimal SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application; the system further includes (d) a detector device that monitors a light-based response from the fluorescent target proteins in the presence of the compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target proteins are disposed, thereby tracking the positions of the fluorescent target proteins, and (ii) track the movement of individual fluorescent target proteins; the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target proteins in the presence of the compound compared to the absence of the compound.

[0149] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include the use of a microscope system configured to identify biological interactions between a compound and a fluorescent target protein in living cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample are positioned within a field of view of the sample plane; and the system further including: (d) a microscope system configured to identify a biological interaction between a compound and a fluorescent target protein in living cells; The system further includes a detector device that monitors a light-based response from the fluorescent target protein, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is disposed, thereby tracking the position of the fluorescent target protein, and (ii) track the movement of individual fluorescent target proteins, wherein the average change in the movement of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline movement under DMSO conditions, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining the change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0150] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include the use of a microscope system configured to identify biological interactions between a compound and a fluorescent target protein in living cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, the plurality of fluorescent target proteins in the sample being disposed in the plane of the sample; and (d) an objective lens for focusing the light beam onto the sample in a plane of the sample, the plurality of fluorescent target proteins in the sample being disposed in the plane of the sample. The system further includes a detector device that monitors the response, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target proteins are located, thereby tracking the position of the fluorescent target proteins, and (ii) track the movement of individual fluorescent target proteins, wherein the tracking includes detecting fluorescence from a plurality of fluorescent target proteins within a field of view of the sample surface at a rate of about 10,000 to about 18,000 per day per system, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target proteins in the presence of the compound compared to the absence of the compound.

[0151] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow can include the use of a microscope system configured to identify biological interactions between compounds and fluorescent target proteins in live cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, the plurality of fluorescent target proteins in the sample being disposed in the plane of the sample; and (d) an objective lens for focusing the light beam onto the sample in a plane of the sample. and a detector device for monitoring a light-based response from the fluorescent target protein in the presence of the compound, the detector device being configured to (i) block light received from a light source other than a sample plane where the fluorescent target protein is disposed, thereby tracking the position of the fluorescent target protein, (ii) track the movement of individual fluorescent target proteins, and (iii) achieve a Z-factor of greater than 0.5 based on a single field of view; the system further comprising: (e) a memory; and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0152] 4.2.OLS htSMT binding In certain embodiments of the OLS htSMT workflow described herein, the system and method utilizes the f seen in the htSMT screening assay. bound Increased residence time (k * off Importantly, both FRAP and htSMT are adapted to distinguish between recovery after exposure to compounds caused by an increase in dwell time (k * off decrease in chromatin binding rate (k * on It is not possible to distinguish between recovery caused by boundThis results in an increase in the number of dwell times. By modifying the SMT acquisition conditions to reduce the illumination intensity and collect longer frame exposures, only immobile proteins form spots. Under these imaging conditions, the distribution of track lengths provides a measure of relative dwell times.

[0153] Exemplary OLS htSMT combined workflows include the following individual strategies and combinations of the following strategies that combine two or more strategic requirements: For example, but not by way of limitation, the disclosed workflow includes both illuminating a field of view in a sample plane disposed within a sample with a light beam to produce fluorescence by a subset of fluorescent target proteins in live cells, the subset of fluorescent target proteins being present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, and illuminating a field of view in a sample plane disposed within the sample with a light beam to produce fluorescence by a subset of fluorescent target proteins in live cells, the subset of fluorescent target proteins including proteins in the range of about 1,000 to about 1,000,000 proteins. Similarly, illuminating the sample plane to illuminate about 30 to about 80 live cells per FOV and / or generating fluorescence from about 1,000 to about 1,000,000 proteins can be combined with any of the other strategic requirements disclosed herein, such as determining that the average change in fluorescent target protein motion in the presence of a compound is about 5% to about 10% compared to baseline motion under DMSO conditions, detecting fluorescence from multiple fluorescent target proteins within the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system, and achieving a Z-factor greater than 0.5 based on a single field of view.

[0154] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in live cells is a K offand determining whether the compound reduces the activity of the compound, the workflow including (a) contacting a sample including a population of live cells with a compound, the live cells including a fluorescent target protein, and the workflow further including (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking including (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by at least a subset of the target fluorescent proteins in the live cells, the subset of fluorescent target proteins being present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, depending on the particular cell type used, e.g. For example, for U2OS cells, the range is about 30 to about 40 per FOV, while for HCT116 cells, the range is about 50 to about 80, taking into account the difference in area. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method being adapted to selectively detect localized fluorescence. The workflow further includes (c) determining a change in the movement of the fluorescent target protein in the presence of a compound, where an increase in the signal detected from the fluorescent target protein in the presence of the compound compared to the signal from the fluorescent target protein in the absence of the compound indicates that the compound has a K off This indicates that it induces a decrease in

[0155] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in live cells is a K offand determining whether a compound reduces the expression level of the protein of interest, the workflow comprising: (a) contacting a sample comprising a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking comprising: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by at least a subset of the target fluorescent proteins in the live cells; the subset of fluorescent target proteins comprising a range of about 1,000 to about 1,000,000 proteins; and the number of proteins in the subset is determined based on the expression level of the protein of interest and a robust method for detecting the number of proteins in the subset. and a dye concentration deemed appropriate for labeling the subset proteins for SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application; the tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method being adapted to selectively detect localized fluorescence; the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound, wherein an increase in the signal detected from the fluorescent target proteins in the presence of the compound compared to the signal of the fluorescent target proteins in the absence of the compound indicates that the compound has a K off This indicates that it induces a decrease in

[0156] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in live cells is a K offthe workflow may include determining whether the compound reduces the K of the fluorescent target protein, the workflow including (a) contacting a sample including a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking including (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to cause fluorescence by at least a subset of the target fluorescent proteins in the live cells; and (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method adapted to selectively detect localized fluorescence; and (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, the average change in the movement of the fluorescent target protein in the presence of the compound being about 5% to about 10% compared to baseline movement under DMSO conditions; and an increase in the signal detected from the fluorescent target protein in the presence of the compound compared to the signal of the fluorescent target protein in the absence of the compound, indicating that the compound reduces the K of the fluorescent target protein. off This indicates that it induces a decrease in

[0157] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in live cells is a K offthe workflow may include determining whether a compound reduces K of the fluorescent target protein, the workflow including (a) contacting a sample including a population of live cells with a compound, the live cells including a fluorescent target protein, the workflow further including (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking including (i) illuminating with a light beam a field of view in a sample plane disposed within the sample so as to cause fluorescence by at least a subset of the target fluorescent proteins in the live cells, and (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method adapted to selectively detect localized fluorescence, the tracking further including (iii) achieving a Z-factor of greater than 0.5 based on a single field of view, the workflow further including (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein an increase in a signal detected from the fluorescent target protein in the presence of the compound compared to a signal from the fluorescent target protein in the absence of the compound indicates that the compound reduces K of the fluorescent target protein. off This indicates that it induces a decrease in

[0158] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow measures the capacity of a compound to induce a change in binding of a fluorescent target protein in live cells by determining whether the compound has a K offThe workflow can include determining whether a compound reduces the activity of a cell of interest by determining whether the compound reduces the activity of the cell of interest, the workflow comprising: (a) contacting a sample comprising a population of live cells with a compound, the live cells comprising a fluorescent target protein; the workflow further comprising: (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking comprising: (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to induce fluorescence by at least a subset of the target fluorescent proteins in the live cells, the subset of fluorescent target proteins being detected by about 30 to about 80 live cells illuminated in the field of view in the sample plane, depending on the particular cell type used. For example, in the case of U2OS cells, the range is about 30 to about 40 per FOV, while in the case of HCT116 cells, the range is about 50 to about 80, taking into account the difference in area. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method being adapted to selectively detect localized fluorescence. The workflow further includes (c) determining a change in the movement of the fluorescent target protein in the presence of a compound, where an increase in the signal detected from the fluorescent target protein in the presence of the compound compared to the signal of the fluorescent target protein in the absence of the compound indicates that the compound has a K off This indicates that increasing the dose induces a decrease in drug metabolism due to increased residence time.

[0159] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow measures the capacity of a compound to induce a change in binding of a fluorescent target protein in live cells by determining whether the compound has a K offdetermining whether a compound reduces the expression level of the protein of interest, the workflow comprising: (a) contacting a sample comprising a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking comprising: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by at least a subset of the target fluorescent proteins in the live cells, the subset of fluorescent target proteins comprising a range of about 1,000 to about 1,000,000 proteins, the number of proteins in the subset being indicative of an expression level of the protein of interest. and a dye concentration deemed appropriate for labeling the subset proteins for robust SMT, both of which can be calculated and / or configured by one skilled in the art based on the disclosure of the present application; the tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method being adapted to selectively detect localized fluorescence; the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound, where an increase in the signal detected from the fluorescent target proteins in the presence of the compound compared to the signal of the fluorescent target proteins in the absence of the compound indicates that the compound has a K off These studies have shown that increasing the dose induces a decrease in the drug's metabolism, and in certain cases, increases in the drug's residence time, leading to a decrease in drug metabolism.

[0160] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow measures the capacity of a compound to induce a change in binding of a fluorescent target protein in live cells by determining whether the compound has a K offthe workflow may include (a) contacting a sample containing a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking including (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to cause fluorescence by at least a subset of the target fluorescent proteins in the live cells; and (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method adapted to selectively detect localized fluorescence; and (c) determining a change in the movement of the fluorescent target proteins in the presence of the compound, the average change in the movement of the fluorescent target proteins in the presence of the compound being about 5% to about 10% compared to baseline movement under DMSO conditions; and an increase in the signal detected from the fluorescent target proteins in the presence of the compound compared to the signal of the fluorescent target proteins in the absence of the compound, which indicates that the compound has a K of the fluorescent target proteins. off This indicates that increasing the dose induces a decrease in drug metabolism due to increased residence time.

[0161] In certain embodiments of the OLS htSMT binding workflow described herein, the workflow measures the capacity of a compound to induce a change in binding of a fluorescent target protein in live cells by determining whether the compound has a K offthe workflow may include (a) contacting a sample comprising a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of individual fluorescent target proteins in a plurality of cells in the sample, the tracking including (i) illuminating with a light beam a field of view in a sample plane disposed within the sample so as to cause fluorescence by at least a subset of the target fluorescent proteins in the live cells; and (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, the method adapted to selectively detect localized fluorescence; the tracking further including (iii) achieving a Z-factor of greater than 0.5 based on a single field of view; and the workflow may further include (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein an increase in a signal detected from the fluorescent target protein in the presence of the compound compared to a signal from the fluorescent target protein in the absence of the compound indicates that the compound is a signal that corresponds to a K for the fluorescent target protein. off These results demonstrate that the drug induces a decrease in the amount of steroid hormone and, in some cases, an increase in dosage due to increased residence time leading to decreased drug metabolism.

[0162] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in a cell is a K of the fluorescently labeled target. offThe method can include using a microscope system configured to determine whether a fluorescent target protein reduces a cell's activity, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a sample plane, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view in the sample plane, the subset of fluorescent target proteins being present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, depending on the particular cell type used; e.g., for U2OS cells, the range is about 30 to about 40 per FOV, while for HCT116 cells, the range is about 30 to about 40 per FOV. In the case of (a), the range is about 50 to about 80, taking into account differences in area; the system further includes (d) a detector device that monitors a light-based response from the fluorescent target protein in the presence of a compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is located, thereby tracking the position of the fluorescent target protein, and (ii) track the movement of each fluorescent target protein, wherein the tracking is adapted to selectively detect localized fluorescence relative to dynamic fluorescence; the system further includes (e) a memory; and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0163] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in a cell is a K of the fluorescently labeled target. offthe system may include use of a microscope system configured to determine whether a protein of interest is reduced, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample; a subset of the fluorescent target proteins in the sample is positioned within a field of view of the sample plane; the subset of fluorescent target proteins includes proteins in the range of about 1,000 to about 1,000,000; the number of proteins in the subset depends on the expression level of the protein of interest and a dye concentration deemed appropriate for labeling the subset proteins for robust SMT; Both of these can be calculated and / or configured by one skilled in the art based on the disclosure of the present application, and the system further includes (d) a detector device that monitors a light-based response from the fluorescent target protein in the presence of the compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is disposed, thereby tracking the position of the fluorescent target protein, and (ii) track the movement of individual fluorescent target proteins, wherein the tracking is adapted to selectively detect localized fluorescence relative to dynamic fluorescence, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0164] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in a cell is a K of the fluorescently labeled target. offThe present invention can include the use of a microscope system configured to determine whether a compound reduces a light-based response from the fluorescent target proteins in the sample, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view of the sample plane; and the system further including: (d) a detector device for monitoring the light-based response from the fluorescent target proteins in the presence of a compound, the detector device detecting (i) the fluorescent target proteins; (i) blocking light received from a light source other than the sample surface where the protein is located, thereby tracking the position of the fluorescent target protein; and (ii) tracking the movement of individual fluorescent target proteins, wherein the tracking is adapted to selectively detect localized fluorescence relative to dynamic fluorescence, and the average change in the movement of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline movement under DMSO conditions. The system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining the change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0165] In certain embodiments of the OLS htSMT screening workflow described herein, the workflow determines whether a compound that induces a change in binding of a fluorescent target protein in a cell is a K of the fluorescently labeled target. offThe present invention can include the use of a microscope system configured to determine whether a compound reduces a light-based response from the fluorescent target proteins in the sample, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view of the sample plane; and the system further including: (d) a detector device for monitoring the light-based response from the fluorescent target proteins in the presence of a compound, the detector device detecting (i) the fluorescent target proteins; The system is configured to (i) block light received from a light source other than the sample surface where the optical target proteins are located, thereby tracking the position of the fluorescent target proteins; and (ii) track the movement of individual fluorescent target proteins, wherein the tracking is adapted to selectively detect localized fluorescence over dynamic fluorescence, and the detector device is further configured to (iii) achieve a Z-factor of greater than 0.5 based on a single field of view, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining changes in the movement of the fluorescent target proteins in the presence of a compound compared to the absence of the compound.

[0166] 4.3.OLS Kinetic SMT Because SMT can identify the rate of biological interaction between a compound and a target, it can be used to distinguish between direct and indirect effects on target activity, among other parameters. Given the live-cell setting of SMT, a data collection mode (kinetic SMT or kSMT) can be configured that allows protein movement after compound addition to be measured at set intervals to determine the rate of biological interaction between the compound and the target.

[0167] Exemplary OLS kinetic SMT workflows include the following individual strategies and combinations of the following strategies that combine two or more strategic requirements: For example, but not by way of limitation, the disclosed workflow includes both illuminating a field of view in a sample plane disposed within a sample with a light beam to induce fluorescence by a subset of fluorescent target proteins in live cells, the subset of fluorescent target proteins being present in about 30 to about 80 live cells illuminated within the field of view in the sample plane, and illuminating a field of view in a sample plane disposed within the sample with a light beam to induce fluorescence by a subset of fluorescent target proteins in live cells, the subset of fluorescent target proteins comprising a range of about 1,000 to about 1,000,000 proteins. Similarly, illuminating the sample plane to illuminate about 30 to about 80 live cells per FOV and / or generating fluorescence from about 1,000 to about 1,000,000 proteins can be combined with any of the other strategic requirements disclosed herein, such as determining that the average change in fluorescent target protein motion in the presence of a compound is about 5% to about 10% compared to baseline motion under DMSO conditions, detecting fluorescence from multiple fluorescent target proteins within the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system, and achieving a Z-factor greater than 0.5 based on a single field of view.

[0168] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include identifying an incidence of biological interactions between a compound and a target in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including a fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample at multiple time points, the tracking including: (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to induce fluorescence by a subset of the target fluorescent proteins in the live cells; The protein subset is present in approximately 30 to approximately 80 illuminated live cells within the field of view of the sample plane, depending on the specific cell type used. For example, for U2OS cells, the range is approximately 30 to approximately 40 per FOV, while for HCT116 cells, the range is approximately 50 to approximately 80, taking into account differences in area. The tracking further includes (ii) detecting fluorescence from multiple fluorescent target proteins within the field of view of the sample plane via a detector device. The workflow further includes (c) determining changes in the movement of the fluorescent target proteins in the presence of a compound, where the rate at which changes in the movement of the fluorescent target proteins occur in the presence of the compound indicates the rate of biological interaction between the compound and the target.

[0169] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include identifying an incidence of biological interactions between a compound and a target in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including a fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample at multiple time points, the tracking including: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to induce fluorescence by a subset of the target fluorescent proteins in the live cells; the subset of fluorescent target proteins comprising at least about 1000 individual fluorescent target proteins. The subset includes proteins in the range of 1 to about 1,000,000, where the number of proteins in the subset depends on the expression level of the protein of interest and the dye concentration deemed appropriate for labeling the subset proteins for robust SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application. The tracking further includes (ii) detecting fluorescence from multiple fluorescent target proteins within a field of view of the sample plane via a detector device, and the workflow further includes (c) determining a change in the motion of the fluorescent target proteins in the presence of a compound, where the rate at which the change in motion of the fluorescent target proteins occurs in the presence of the compound indicates the rate of occurrence of a biological interaction between the compound and the target.

[0170] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include identifying a rate of biological interaction between a compound and a target in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells including a fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample at multiple time points, the tracking including: (i) illuminating a field of view in a sample plane disposed within the sample with a light beam to induce fluorescence by a subset of the target fluorescent proteins in the live cells; and (ii) detecting fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane via a detector device; and (c) determining a change in the movement of the fluorescent target proteins in the presence of the compound, wherein the average change in the movement of the fluorescent target proteins under DMSO conditions is about 5% to about 10% compared to baseline movement in the absence of the compound; and the rate at which the change in the movement of the fluorescent target proteins in the presence of the compound indicates a rate of biological interaction between the compound and the target.

[0171] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include identifying incidences of compound-target biological interactions among direct and indirect biological interactions between a compound and a fluorescent target protein in live cells, the workflow including (a) contacting a sample including a population of live cells with a compound, the live cells including the fluorescent target protein, the workflow further including (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample at multiple time points, the tracking including (i) detecting fluorescence by a subset of the target fluorescent proteins in the live cells. (ii) illuminating a field of view in a sample plane disposed within the sample with a light beam to induce a signal; and (iii) detecting fluorescence from a plurality of fluorescent target proteins within the field of view of the sample plane via a detector device; (iii) the tracking step includes detecting the fluorescence from the plurality of fluorescent target proteins within the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system; and the workflow further includes (c) determining a change in the movement of the fluorescent target proteins in the presence of a compound, wherein the rate at which the change in the movement of the fluorescent target proteins occurs in the presence of the compound indicates a rate of occurrence of a biological interaction between the compound and the target.

[0172] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include identifying a rate of occurrence of a biological interaction between a compound and a target in live cells, the workflow including: (a) contacting a sample including a population of live cells with a compound, the live cells comprising a fluorescent target protein; and (b) tracking the movement of a plurality of individual fluorescent target proteins in a plurality of live cells in the sample at multiple time points, the tracking including: (i) illuminating with a light beam a field of view in a sample plane disposed within the sample to cause fluorescence by a subset of the target fluorescent proteins in the live cells; and (ii) detecting, via a detector device, fluorescence from the plurality of fluorescent target proteins in the field of view in the sample plane, wherein detection based on a single field of view is associated with a Z-factor of greater than 0.5; and (c) determining a change in the movement of the fluorescent target protein in the presence of the compound, wherein a rate at which the change in the movement of the fluorescent target protein occurs in the presence of the compound indicates a rate of occurrence of a biological interaction between the compound and the target.

[0173] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include determining a dose of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; and (b) tracking the movement of individual target fluorescent proteins in a plurality of live cells of the sample at multiple time points, the tracking including: (i) illuminating with a light beam a field of view in a sample plane located within the sample to induce fluorescence by at least a subset of the fluorescent target proteins in the live cells; The subset of target proteins is present in approximately 30 to approximately 80 live cells illuminated within the field of view of the sample plane, depending on the specific cell type used. For example, for U2OS cells, the range is approximately 30 to approximately 40 per FOV, while for HCT116 cells, the range is approximately 50 to approximately 80 per FOV, taking into account differences in area. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device. The workflow further includes (c) determining the rate at which changes in the movement of the fluorescent target proteins occur in the presence of the compound; and (d) repeating steps (b) to (c) for each of multiple samples over a range of compound concentrations, wherein the rate at which changes in the movement of the fluorescent target proteins occur in the presence of the compound indicates the rate of occurrence of biological interactions between the compound and the target.

[0174] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include determining a dose of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow comprising: (a) contacting a plurality of samples with a compound, (i) each sample comprising a population of live cells; (ii) the live cells comprising a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further comprising: (b) tracking the movement of individual target fluorescent proteins in a plurality of live cells of the sample at multiple time points, the tracking comprising: (i) illuminating with a light beam a field of view in a sample plane located within the sample to induce fluorescence by at least a subset of the fluorescent target proteins in the live cells; The subset may include proteins ranging from about 1,000 to about 1,000,000, where the number of proteins in the subset depends on the expression level of the protein of interest and the dye concentration deemed appropriate for labeling the subset proteins for robust SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application. The tracking further includes (ii) detecting fluorescence from one or more of the fluorescent target proteins in the sample plane via a detector device, where the workflow further includes (c) determining the rate at which a change in the motion of the fluorescent target protein occurs in the presence of the compound; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, where the rate at which a change in the motion of the fluorescent target protein occurs in the presence of the compound indicates the rate of occurrence of a biological interaction between the compound and the target.

[0175] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include determining a dose response of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further including: (b) tracking the movement of individual target fluorescent proteins in a plurality of live cells of the samples at multiple time points; and (ii) detecting fluorescence from one or more fluorescent target proteins in the sample plane via a detector device; the workflow further includes (c) determining a rate at which a change in the motion of the fluorescent target protein occurs in the presence of the compound, wherein the average change in the motion of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline motion under DMSO conditions; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, wherein the rate at which the change in the motion of the fluorescent target protein occurs in the presence of the compound indicates a rate of occurrence of a biological interaction between the compound and the target.

[0176] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include determining a dose response of a compound to induce a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; the workflow further including: (b) tracking the movement of individual target fluorescent proteins in a plurality of live cells of the samples at multiple time points; (ii) illuminating with a light beam a field of view in a sample plane disposed within a filter; and (ii) detecting, via a detector device, fluorescence from one or more of the fluorescent target proteins in the sample plane. (iii) said tracking includes detecting fluorescence from a plurality of fluorescent target proteins in the field of view of the sample plane at a rate of about 10,000 to about 18,000 per day per system. The workflow further includes (c) determining a rate at which a change in the motion of the fluorescent target protein occurs in the presence of a compound; and (d) repeating steps (b)-(c) for each of a plurality of samples over a range of compound concentrations, wherein the rate at which a change in the motion of the fluorescent target protein occurs in the presence of the compound indicates a rate of occurrence of a biological interaction between the compound and the target.

[0177] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include determining a dose of a compound that induces a change in the movement of a fluorescent target protein in a live cell, the workflow including: (a) contacting a plurality of samples with a compound, (i) each sample including a population of live cells; (ii) the live cells including a fluorescent target protein; and (iii) the plurality of samples being contacted with different concentrations of the compound over a range of compound concentrations; and the workflow further including: (b) tracking the movement of individual target fluorescent proteins in a plurality of live cells of the samples at a plurality of time points, wherein the tracking includes: (i) determining a change in the movement of at least one fluorescent target protein in the live cells. (ii) illuminating with a light beam a field of view in a sample plane disposed within the sample so as to cause fluorescence from at least a subset of the fluorescent target proteins; and (ii) detecting via a detector device fluorescence from one or more of the fluorescent target proteins within the field of view in the sample plane, wherein detection based on a single field of view is associated with a Z-factor of greater than 0.5; and (c) determining a rate at which a change in the motion of the fluorescent target proteins occurs in the presence of the compound; and (d) repeating steps (b)-(c) for each of a plurality of samples across a range of compound concentrations, wherein the rate at which a change in the motion of the fluorescent target proteins occurs in the presence of the compound indicates a rate of occurrence of a biological interaction between the compound and the target.

[0178] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include the use of a microscope system configured to determine the incidence of biological interactions between a compound and a target in live cells, the system including: (a) a stage supporting a sample, the sample including a population of live cells, the live cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample is positioned within a field of view of the sample plane, the subset of fluorescent target proteins being detected by about 30 to about 80 live cells illuminated within the field of view of the sample plane, depending on the particular cell type used. For example, in the case of U2OS cells, the range is about 30 to about 40 per FOV, while in the case of HCT116 cells, the range is about 50 to about 80, taking into account the difference in area. The system further includes (d) a detector device that monitors a light-based response from the fluorescent target protein in the presence of a compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is located, thereby tracking the position of the fluorescent target protein at multiple time points, and (ii) track the movement of individual fluorescent target proteins. The system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, the processor being capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0179] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include use of a microscopy system configured to determine the incidence of biological interactions between a compound and a target in live cells, the system including: (a) a stage supporting a sample, the sample including a population of cells, the cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens that focuses the light beam onto the sample in a plane of the sample; a subset of the fluorescent target proteins in the sample is positioned within a field of view in the sample plane; the subset of fluorescent target proteins includes a range of about 1,000 to about 1,000,000 proteins; and the number of proteins in the subset is determined based on the expression level of the protein of interest and the number of proteins in the subset. and a dye concentration deemed appropriate for labeling the subset proteins for robust SMT, both of which can be calculated and / or configured by one of ordinary skill in the art based on the disclosure of the present application; the system further includes (d) a detector device that monitors a light-based response from the fluorescent target proteins in the presence of the compound, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target proteins are located, thereby tracking the positions of the fluorescent target proteins at multiple time points, and (ii) track the movement of individual fluorescent target proteins; the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, the processor being capable of determining a change in the movement of the fluorescent target proteins in the presence of the compound compared to the absence of the compound.

[0180] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include the use of a microscope system configured to determine the incidence of biological interactions between a compound and a target in live cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens for focusing the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample are positioned within a field of view of the sample plane; and the system further including: (d) a microscope system configured to determine the incidence of biological interactions between a compound and a target in live cells, the system including: (a) a stage for supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; The system further includes a detector device that monitors a light-based response from the target protein, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is disposed, thereby tracking the position of the fluorescent target protein at multiple time points, and (ii) track the movement of each fluorescent target protein, wherein an average change in the movement of the fluorescent target protein in the presence of the compound is about 5% to about 10% compared to baseline movement under DMSO conditions, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining the change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0181] In certain embodiments of the OLS kinetic htSMT combined workflow described herein, the workflow can include the use of a microscope system configured to determine the incidence of a biological interaction between a compound and a target in live cells, the system including: (a) a stage supporting a sample, the sample including a population of cells, the cells including a fluorescent target protein; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens focusing the light beam onto the sample in a plane of the sample, the plurality of fluorescent target proteins in the sample being disposed in the plane of the sample; and (d) an objective lens focusing the light beam onto the sample in a plane of the sample, the plurality of fluorescent target proteins in the sample being disposed in the plane of the sample. The system further includes a detector device for monitoring a response of the fluorescent target protein, the detector device being configured to (i) block light received from a light source other than the sample surface on which the fluorescent target protein is disposed, thereby tracking the position of the fluorescent target protein at multiple time points, and (ii) track the movement of individual fluorescent target proteins, wherein the tracking includes detecting fluorescence from the plurality of fluorescent target proteins within a field of view of the sample surface at a rate of about 10,000 to about 18,000 per day per system, and the system further includes (e) a memory and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0182] In certain embodiments of the OLS kinetic htSMT binding workflow described herein, the workflow can include the use of a microscope system configured to determine the incidence of biological interactions between a compound and a target in live cells, the system including: (a) a stage supporting a sample, the sample including a population of live cells, the live cells including fluorescent target proteins; the system further including: (b) a light source emitting a light beam capable of inducing a light-based response from a plurality of fluorescent target proteins in the sample; and (c) an objective lens that focuses the light beam onto the sample in a plane of the sample, wherein a subset of the fluorescent target proteins in the sample are positioned within a field of view of the sample plane; and the system further including: (a) a microscope stage configured to support a sample; (b) a microscope stage configured to support a sample; d) a detector device that monitors a light-based response from the fluorescent target protein in the presence of the compound, the detector device being configured to (i) block light received from a light source other than the sample plane where the fluorescent target protein is located, thereby tracking the position of the fluorescent target protein at multiple time points, (ii) track the movement of individual fluorescent target proteins, and (iii) achieve a Z-factor of greater than 0.5 based on a single field of view; the system further comprising: (e) a memory; and (f) a processor in communication with the memory and the detector device, wherein the processor is capable of determining a change in the movement of the fluorescent target protein in the presence of the compound compared to the absence of the compound.

[0183] 5. Working Example Example 1: Optical line-scan high-throughput single-molecule tracking A. Preface This example describes an industrial-scale OLS htSMT technique, a system incorporating such an OLS htSMT technique, hardware and software associated with such an OLS htSMT technique, and methods for using such an OLS htSMT technique. For example, the OLS htSMT technique described herein is capable of measuring protein movement in millions of cells per day. The OLS htSMT technique described herein demonstrates specific, robust, and reproducible results. The OLS htSMT technique described herein can be used for a variety of applications, including, but not limited to, traditional drug discovery activities such as screening compound libraries and elucidating SAR. Importantly, the OLS htSMT technique described herein can be used to characterize the contributions of both known and novel pathways to interaction networks, such as protein signaling interaction networks.

[0184] B. Results a. Creation and verification of htSMT system We developed a robotic system capable of handling reagents, collecting high-quality, high-speed SMT image series, and processing the time-ordered raw images to generate molecular trajectories and extract biologically interesting features within defined cellular compartments (Figure 1). The FOV between the HILO-based approach (center image) and the OLS-based approach (right image) is shown, along with a comparison of spatial heterogeneity across the FOV of each approach.

[0185] C. Method a.Cell line U2OS cells (ATCC Cat. No. HTB-96) can be grown in DMEM (Cat. No. 1056601, Gibco DMEM, High Glucose, GlutaMAX Supplement, Thermo Fisher) supplemented with 10% fetal bovine serum (Cat. No. 16000044, Thermo Fisher) and 1% penicillin-strep (Cat. No. 15140122, Thermo Fisher), maintained in a humidified 37°C incubator with 5% CO , and subcultured approximately every 2–3 days.

[0186] b. HaloTag-expressing cell line For specific target-HaloTag fusions, mammalian expression vectors containing the appropriate fusion gene under the control of the weak L30 promoter and containing a neomycin resistance marker can be transfected into U2OS cells at 70% confluence using FuGENE6 (Cat. No. E2691, Promega). Transfected cells can be selected with 500 μg / mL G418 (Cat. No. 10131027, Thermo Fisher Scientific) and then clonally isolated. Clones expressing the desired fusion gene can be identified by first selecting with 100 nM JF. 549 -HTL (Cat. No. GA1110, Promega) and 50 nM Hoechst 33342 staining, JF 549 This can be determined by identifying clones with the expected distribution of signals. Many clones can then be tested using SMT conditions for response to a control compound, and the most homogeneous clones can then be expanded for further testing.

[0187] C. Western blot Cells can be grown under the same conditions as described above. 1.5 x 106 cells can be seeded per well in DMEM medium in a 6-well plate and cultured overnight, followed by compound treatment (DMSO or 100 nM fulvestrant) for 24 hours the following day. Cells can then be lysed in 200 μL of 1X Cell Lysis Buffer (Cat. No. 9803, Cell Signaling). Protein lysate concentrations can then be determined using a BCA Protein Assay Kit (Cat. No. 23225, Pierce™ BCA Protein Assay Kit) according to the manufacturer's instructions. Capillary Western immunoassays can then be performed using Jess Protein Simple according to the manufacturer's instructions (Protein Simple, USA). Anti-target antibody levels can be normalized to the loading control β-tubulin (1:100, NC0244815LI-COR92642213, Thermo Fisher Scientific). Peaks can be analyzed using Compass software (Protein Simple, USA).

[0188] d. OLS single molecule tracking sample preparation Cells can then be seeded into tissue-culture-treated 384-well glass-bottom plates at 4,500–6,000 cells per well. The seeded cells can then be incubated overnight at 37°C and 5% CO2 to allow for attachment. For all SMT experiments, cells were incubated with 5–100 pM JF. 549Cells can be incubated with -HTL (catalog no. GA1110, Promega) and 50 nM Hoechst 33342 in complete medium for 1 hour. Cells are then washed three times with DPBS and twice with imaging medium. The imaging medium is fluoroBrite DMEM medium (catalog no. A1896701, Thermo Fisher Scientific) supplemented with GlutaMAX (catalog no. 35050079, Thermo Fisher Scientific) and the same serum and antibiotics as the growth medium. If appropriate, compounds can be serially diluted in Echo-certified 384-well low-dead-volume source microplates (0018544, Beckman Coulter) to generate dose-titrated source material. Compounds can be administered at a final dilution of 1:1000 in cell culture medium. Each compound dose can be replicated at least three times per plate, with up to three plate replicates prepared consecutively. 20 DMSO control wells and two no-dye control wells can be randomized across each plate. Compounds are allowed to incubate at 37° C. for 1 hour before image acquisition.

[0189] e. Image acquisition Unless otherwise noted, all image acquisition using SMT was performed using a custom-built microscope, motorized stage, stage-top environmental chamber, quad-band filter cube (Chroma), and a custom-built laser engine with wavelengths of 405 nm and 561 nm at the back focal plane of the objective. Fluorescence emission was passed through a high-speed filter wheel (Finger Lakes Instruments) and collected by a backlit CMOS camera (Hammamatsu Orca Fusion, running in light sheet mode). Images were acquired with a 60x 1.27NA water-immersion objective (Nikon). The environmental chamber was set at 37°C, 95% humidity, and 5% CO2. In a specific embodiment, the exposure time for each pixel was 400 microseconds, and recording the entire region of interest (ROI) took a total of 9 milliseconds. The galvanometer scanner position was then reset within 1 millisecond, e.g., with the laser turned off, and the next image was recorded. In such an embodiment, 100 frames per second can be recorded. Additionally or alternatively, a second setup using a smaller ROI can be employed to record 200 frames per second with the same 400 microsecond / pixel exposure and 4 millisecond image recording time. Additionally or alternatively, galvo reset can be performed more quickly.

[0190] f. Image analysis Image acquisition yields one JF per field of view 549 A video and one Hoechst were generated. JF 549 The video is from individual JF 549It can be used to track molecular movement, and Hoechst videos can be used for nuclear segmentation. Tracking can be achieved in three sequential steps using a combination of existing methods: detection, subpixel localization, and linking. Briefly, spots can be detected using a generalized log-likelihood ratio detector. After detection, starting from an initial guess obtained by a radial symmetry method, the estimated location of each emitter can be refined to subpixel resolution using Levenberg-Marquardt fitting with a unified 2D Gaussian spot model. Detected spots can be linked into trajectories using a custom modification of a hill-climbing algorithm. The same detection, subpixel localization, and linking setup can be used for all videos.

[0191] For nuclear segmentation, all frames of the Hoechst movie can be averaged to generate an average projection. This average projection can then be segmented with a neural network trained on human-labeled nuclei. Each spot can then be assigned to at most one nucleus using its subpixel coordinates.

[0192] To recover migration information from trajectories, state arrays can be used. For example, a Bayesian inference approach can be used using the "RBME" likelihood function and a grid of 100 diffusion coefficients ranging from 0.01 to 100.0 μm²s-1 and 31 localization error magnitudes ranging from 0.02 to 0.08 μm. After inference, the localization error can be minimized to obtain a one-dimensional distribution across the diffusion coefficients for each field of view. For single-cell analysis, for example, SMT and nuclear segmentation can be performed on a mixture of U2OS cells carrying H2B-HaloTag, HaloTag-CaaX, or free HaloTag. The marginal likelihood of each set of 100 diffusion coefficients for the set of trajectories within each segmented nucleus can then be evaluated. These marginal likelihood functions can be clustered using k-means, and the marginal likelihood functions for each cell can be ordered by its cluster index to create a heatmap. The fractional boundary values ​​(f bound To estimate the free diffusion coefficient (D), we can integrate the posterior distribution of the state sequence below 0.1 μm s. free ) can be calculated using the mean of the posterior distribution over 0.1 μm s.

[0193] g. Data Analysis The tracking results from the automated processing pipeline can be analyzed using KNIME or Spotfire (TIBCO). bound or D free Measurements of f can be associated with experiment metadata and aggregated by condition. bound The change in f of each well bound f of DMSO in the same plate boundThe EC50 can be calculated as the difference from the median of the DMSO wells. Wells that contained no cells within the field of view or where the field of view was out of focus can be excluded from further analysis. Compounds can be evaluated for assay interference using the median fluorescence intensity of the tracking channel and excluded if it is more than three standard deviations higher than the median intensity of the DMSO wells. Similarly, plates that fail to clearly separate the active and negative controls or that deviate significantly from the performance of the rest of the screen can be excluded from further analysis. Finally, compounds with a variance more than three standard deviations above the mean compound variance can be excluded from downstream analysis. The Z' coefficient between the active control and DMSO on a plate can be calculated. EC50 values ​​can be calculated in Prism (GraphPad) by first log-transforming the molecule concentrations and then fitting them to a four-parameter logistic curve.

[0194] h. Clustering of active molecules Chemical structure-based clustering can be performed on the molecules identified as active. Molecular frameworks can be calculated as known in the art and as implemented in pipeline pilots. Molecular frameworks can be clustered using functional class fingerprints (FCFP_4) (e.g., a similarity threshold cutoff of 0.3 Tanimoto distance).

[0195] i. Exercise experiment Cells can be seeded into 384-well plates the day before and stained and washed as described above. One well with multiple FOVs per well can be taken as a baseline reading. Compounds can then be added manually or robotically to each well to a final concentration of 100 nM during imaging. Data can then be collected for that well. Pauses can be included between each FOV to ensure the entire imaging plan covers the assay window. bound The change in can be determined for each well relative to t=0.

[0196] For assays lasting 4 hours, the plate can be imaged twice using multiple FOVs at different FOV positions per well to prevent photobleaching from affecting the data.

[0197] j. Dwell time imaging Sample preparation and execution of dwell time imaging experiments can be performed in a similar manner to the single molecule tracking assay described above, with a few exceptions. Samples should be prepared using 1-10 pM JF. 549 Staining can be performed with 50 nM Hoechst 33342 (Promega) for 1 hour. Multiple frames can be collected per field of view by setting the camera integration time to the desired time (milliseconds) and reducing the laser light source to the desired mW at the objective. The laser can be turned on continuously during image acquisition. Compound incubation can range from 1 to 4 hours.

[0198] k. Residence time analysis Image processing, including spot detection, localization, and tracking reconnection, may be performed using the same methods described above. Because dwell-time imaging selectively tracks slow-diffusing molecules, individual localizations can be limited to the maximum displacement distance of individual jump reconnections. The set of trajectories for each field of view is binned into a 1-CDF distribution as described above and fitted with a biexponential decay model.

number

[0199] l. Fluorescence recovery after photobleaching Images can be acquired with a custom-built OLS microscope using a Spectra Light Engine RS-232, as described herein. Stimulation can be performed directly using a mini-scanner coupled with a Coherent OBIS 561 nm 100 mW laser. All imaging can be performed using a 60x 1.27 NA water immersion objective (Nikon). All experiments can be performed at 37°C. For FRAP experiments, cells are seeded in 384-well plates the day before and treated with 50 nM HTL-JF. 549 The cells can be labeled with ATP and washed as described above. Compounds can be added to a final concentration of 100 nM before imaging. Pre-bleaching images can then be obtained by averaging 10 consecutive images. Next, 8-10 regions can be bleached (two background, six-eight cells), and two regions within the cells can be left unbleached. The bleached regions are then bleached at 10% power without scanning. For the next 30 seconds, images can be acquired every 200 ms, then every 1 second for 2 minutes. The background-subtracted average intensity can be measured over time in the region of interest and normalized to the average fluorescence in the baseline image, which can then be normalized to the unbleached region to account for readout-induced photobleaching of the fluorophore. For three biological experiments, data from multiple cells per experiment can be pooled.

Claims

1. 1. A method comprising: receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells, each image comprising an array of pixels, the method further comprising: Segmenting each image to identify one or more cells; For each image, assigning a cell health score to each pixel of a particular identified cell using at least one machine learning model; calculating, for each identified image, a total cell health score based on the cell health scores assigned to the pixels; providing data characterizing the calculated overall cell health score to a consuming application or process; and The method comprising:

2. 10. The method of claim 1, wherein the microscopic image sequences visualize the fluorescently labeled cellular components while the live cells are exposed to a compound or various environmental conditions.

3. The method of claim 2 , wherein the microscopic image is at field level.

4. The method of claim 3, further comprising calculating the percentage of healthy cells within the field of view.

5. 10. The method of any preceding claim, further comprising staining the live cells using Hoechst dye.

6. 10. The method of any preceding claim, further comprising generating a training data set by selectively exposing said sample of live cells to at least one compound at a range of concentrations.

7. 3. The method of claim 1 or 2, further comprising generating a training data set by selectively exposing samples of said live cells to at least one compound at different concentrations and for different exposure times.

8. 8. The method of claim 6 or 7, wherein the at least one compound comprises staurosporine and / or sorbitol.

9. a training dataset for the at least one machine learning model, staining a sample of live cells with a fluorescent dye; exposing a first subset of the sample of live cells to a concentration of a first compound at a first concentration corresponding to unhealthy cells, wherein the first compound affects cell health; exposing a second subset of the sample of live cells to a concentration of the first compound at a second concentration corresponding to healthy cells; exposing a third subset of the sample of live cells to a concentration of a second compound at a third concentration corresponding to unhealthy cells, wherein the second compound affects cell health; exposing a fourth subset of the sample of live cells to a concentration of the second compound at a fourth concentration corresponding to healthy cells; generating images for each of the first subset, the second subset, the third subset, and the fourth subset of the sample of cells using a microscope; labeling the images of the first and third subsets of the sample of live cells as corresponding to unhealthy cells; labeling the images of the second and fourth subsets of the sample of live cells as corresponding to healthy cells; training the at least one machine learning model with the training dataset; 10. The method of any preceding claim, further comprising generating by:

10. 10. The method of claim 9, further comprising varying the exposure time of the first compound to the first and second subsets of the sample of live cells.

11. 11. The method of claim 10, further comprising varying the exposure time of the second compound to the third and fourth subsets of the sample of live cells.

12. The method of any one of claims 9 to 11, wherein the first compound is staurosporine and the second compound is sorbitol.

13. The method of any of claims 6 to 12, further comprising randomly varying the brightness levels of the images generated for the training data set.

14. 14. The method of any of claims 6 to 13, further comprising manually labeling the subset of images generated for the training dataset at a pixel level as corresponding to one of healthy cells, unhealthy cells, or background.

15. 10. The method of any preceding claim, wherein the at least one machine learning model comprises a pixel-level convolutional neural network.

16. The method of claim 15 , wherein the convolutional neural network comprises a U-Net.

17. The method of claim 16 , further comprising optimizing the U-Net for weighted categorical cross-entropy.

18. The method of any preceding claim, wherein providing the data includes one or more of: visualizing at least a portion of the data characterizing the calculated total cell health score in a graphical user interface; storing at least a portion of the data characterizing the calculated total cell health score in a physically persistent state; loading at least a portion of the data characterizing the calculated total cell health score into memory; or transmitting at least a portion of the data characterizing the generated calculated total cell health score to a remote computing device via a network.

19. determining, for each identified cell, a cell health subtype score for each cell using at least one second machine learning model; providing data characterizing the determined cell health subtype score to a consuming application or process; and 10. The method of any preceding claim, further comprising:

20. 1. A method comprising: receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells, each image comprising an array of pixels, the method further comprising: Segmenting each image to identify one or more cells; For each image, using at least one machine learning model, classifying each pixel as corresponding to a healthy cell, an unhealthy cell, or background; calculating, for each identified cell, a total cell health score based on the classification of the corresponding pixel; providing data characterizing the calculated overall cell health score to a consuming application or process; and The method comprising:

21. 1. A method comprising: receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells, each image comprising an array of pixels, the method further comprising: For each image, classifying each pixel of a particular identified cell as corresponding to a healthy cell or an unhealthy cell using at least one machine learning model, wherein the at least one machine learning model is trained using weak supervision in which training images are labeled at the field level, the method further comprising: calculating, for each identified cell, a total cell health score based on the classified labeled pixels; providing data characterizing the calculated overall cell health score to a consuming application or process; and The method comprising:

23. 1. A method comprising: receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells, each image comprising an array of pixels, the method further comprising: Segmenting each image to identify one or more cells; For each image, assigning a cell health subtype score to each pixel of a particular identified cell using at least one machine learning model; calculating, for each identified image, a total cell health subtype score based on the cell health subtype scores assigned to the pixels; providing data characterizing said calculated total cell health subtype score to a consuming application or process; and The method comprising:

24. 23. The method of claim 22, wherein the cell health subtype score characterizes apoptosis, oxidative stress, and / or inflammation of the corresponding cell.

25. 1. A method comprising: receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells; Segmenting each image to identify one or more cells; determining a cell health score and at least one cell health subtype score for each identified cell using an ensemble of separately trained machine learning models; providing data characterizing said score to a consuming application or process; The method comprising:

26. 1. A system comprising: at least one data processor; a memory storing instructions which, when executed by at least one data processor, cause operations to be performed to implement a method according to any of the preceding claims; The system comprising:

27. A non-transitory computer program product storing instructions which, when executed by at least one data processor forming part of at least one computing device, perform the method according to any of claims 1 to 25.

28. 1. A system comprising: means for receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within living cells, each image comprising an array of pixels, the system further comprising: means for segmenting each image to identify one or more cells; means for assigning, for each image, a cell health score to each pixel of a particular identified cell using at least one machine learning model; means for calculating, for each identified image, a total cell health score based on the cell health scores assigned to the pixels; a means for providing data characterizing said calculated overall cell health score to a consuming application or process; The system comprising:

29. 1. A system comprising: means for receiving a sequence of microscopic images visualizing fluorescently labeled cellular components within live cells; means for determining, for each identified cell, a cell health subtype score for each cell using at least one machine learning model; a means for providing data characterizing said determined cell health subtype score to a consuming application or process; The system comprising:

Images