Characterization and sorting of particle analyzers

Abstract

Non-parametric transforms (e.g., t-distributed random neighbor embedding (tSNE)) may be used to analyze multi-parameter data, such as data derived from flow cytometry or other particle analysis systems and methods. The present invention may include these transformations for dimensionality reduction and subgroup identification (e.g., gating). In essence, non-parametric transformations cannot transform new observations without training new transformations based on the entire data set, including new observations. The functionality parameterizes non-parametric transformations using neural networks, allowing small training data sets to be transformed using non-parametric techniques. The training data set may then be used to generate an accurate parametric model in order to evaluate other events in a manner consistent with the initial event.

Description

[0001] This application is a divisional application, and its original application is a PCT application with application number PCT / US2019 / 049151 and application date August 30, 2019, and entered the Chinese national phase on April 27, 2021, with application number 201980071128.8, and the name is "Characterization and Sorting of Particle Analyzers".

[0002] Cross-references to Related Patent Applications

[0003] This patent application claims priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No. 62 / 724,834, filed on August 30, 2018; the contents of which are incorporated herein by reference. Technical Field

[0004] The present invention relates generally to the field of automated particle evaluation technology, and more particularly to sample analysis and particle characterization methods. Background Art

[0005] Particle analyzers (e.g., flow cytometers and scanning cytometers) are analytical tools that can characterize particles based on electro-optical measurements such as light scattering and fluorescence. In a flow cytometer, for example, particles in a fluid suspension (e.g., molecules, microbeads bound to an analyte, or individual cells) pass through a detection area where they are exposed to excitation light, typically from one or more lasers, and the light scattering and fluorescence properties of the particles are measured. Particles or their components are typically labeled with fluorescent dyes for detection. Various different particles or components can be detected simultaneously by labeling them with fluorescent dyes with different spectral properties. In some embodiments, the analyzer includes multiple photodetectors, one for each scattering parameter to be measured and one or more for each different dye to be detected. For example, some embodiments include a spectral structure in which more than one sensor or detector is used for each dye. The data obtained includes the signals measured for each of the light scattering detector and the fluorescence emission.

[0006] The particle analyzer may further comprise means for recording the measured data and analyzing the data. For example, a computer connected to the detection electronics may be used for data storage and analysis. For example, the data may be stored in a tabular format, wherein each row corresponds to data for one particle and the columns correspond to each measured characteristic. Storing data from a particle analyzer using a standard file format (e.g., the "FCS" file format) facilitates analysis of the data using a separate program and / or machine. Using current analysis methods, the data is typically displayed in the form of a one-dimensional histogram or a two-dimensional (2D) graph for ease of visualization, but other methods may be used to visualize multi-dimensional data.

[0007] For example, parameters measured using flow cytometry typically include light scattered primarily in the forward direction at a narrow angle by the particles at the excitation wavelength (referred to as forward scatter (FSC)); the excitation light scattered by the particles in a direction orthogonal to the excitation laser (referred to as side scatter (SSC)); and light emitted by fluorescent molecules in one or more detectors (used to measure signals within a certain spectral wavelength range), or light emitted by fluorescent dyes detected primarily in the specific detector or detector array. Different cell types can be identified by light scattering characteristics and fluorescence emissions obtained / generated by labeling various cellular proteins or other components with fluorescent dye-labeled antibodies or other fluorescent probes.

[0008] Flow cytometers and scanning cytometers are commercially available from, for example, BD Biosciences (San Jose, CA). Flow cytometry is described in, for example, Landy et al. (eds.), Clinical Flow Cytometry, Annals of the New York Academy of Sciences, Vol. 677 (1993); Bauer et al. (eds.), Clinical Flow Cytometry: Principles and Applications, Williams & Wilkins (1993); Ormerod (ed.), Flow Cytometry: A Practical Approach, Oxford University Press (1994); Jaroszeski et al. (eds.), Flow Cytometry Protocols, Methods in Molecular Biology, Vol. 91, Humana Press (1997); and Shapiro, Practical Flow Cytometry, 4th ed., Wiley-Liss (2003); all of which are incorporated herein by reference. Fluorescence imaging microscopy is described in, for example, Pawley (ed.), Handbook of Biological Confocal Microscopy, 2nd ed., Plenum Press (1989), which is incorporated herein by reference.

[0009] The data obtained by analyzing cells (or other particles) using multicolor flow cytometry is multidimensional, where each cell corresponds to a point in the multidimensional space defined by the measured parameters. Cell populations or particle populations are identified as clusters of points in the data space. Clusters and populations can be manually identified by drawing gates around a population displayed in one or more two-dimensional plots of the data (referred to as "scatter plots" or "dot plots"). Alternatively, clusters can be automatically identified and gates defining the boundaries of the populations can be automatically determined. Examples of methods for automatic gating are shown in the following publications, for example, U.S. Patents Nos. 4,845,653; 5,627,040; 5,739,000; 5,795,727; 5,962,238; 6,014,904; 6,944,338; and U.S. Patent Publication No. 2012 / 0245889, each of which is incorporated herein by reference.

[0010] Flow cytometry is an effective method for analyzing and separating biological particles (e.g., cells and their constituent molecules), and as such, it is widely used in diagnostics and therapeutics. The method utilizes a fluid medium to linearly separate particles, allowing them to align and pass through a detection device. Individual cells can be distinguished based on their location within the fluid medium and the presence of a detectable marker. Therefore, flow cytometry can be used to characterize and generate diagnostic profiles of biological particle populations.

[0011] Separation of biological particles has been achieved by integrating a sorting or collection function into flow cytometers. Particles in the separation stream that have been detected to possess one or more desired characteristics are individually separated from the sample stream through mechanical or electrical separation. This flow sorting method has been used to sort different cell types, separate sperm containing X and Y chromosomes for animal breeding, sort chromosomes for genetic analysis, and isolate specific organisms from complex biological populations.

[0012] Gating is used to help understand and classify the large amount of data that may be generated by a sample. Given the large amount of data presented by a given sample, there is a need to effectively control the graphical display of that data.

[0013] Fluorescence-activated particle sorting (FACS) or cell sorting is a specialized form of flow cytometry. It provides a method for sorting a heterogeneous mixture of particles into one or more containers, sorting one cell at a time based on the specific light scattering and fluorescence properties of each cell. It records the fluorescence signals from individual cells and physically separates the specific cells of interest. The abbreviation FACS is a trademark of and is owned by Becton Dickinson and may be used to refer to equipment that performs fluorescence-activated particle sorting or cell sorting.

[0014] The particle suspension is placed near the center of a narrow, fast-flowing liquid stream. The stream is arranged so that, on average, there are large spacings between particles relative to their diameters when they arrive randomly (Poisson process) at the detection area. The vibration mechanism causes the outflowing fluid medium to be steadily broken up into single droplets, which contain the particles previously characterized in the detection area. The system is usually adjusted so that the probability of more than one particle being present in a droplet is low. If a particle is classified as being to be collected, an electric charge is applied to the flow cell and the outflow stream over a period of time to form one or more droplets and separate them from the stream. These charged droplets then pass through an electrostatic deflection system, which transfers the droplets to a target container based on the charge applied to the droplets.

[0015] The sample may contain thousands, if not millions, of cells. The cells can be sorted to purify the sample to cells of interest. The sorting process typically identifies three types of cells: cells of interest, non-cells of interest, and unidentifiable cells. To sort cells with high purity (e.g., a high concentration of cells of interest), the cell sorter that generates the droplets typically electronically aborts the sorting if the desired cell is too close to another undesirable cell, thereby reducing contamination of the sorted population by inadvertent inclusion of undesirable particles in droplets containing particles of interest. Summary of the Invention

[0016] In one innovative aspect, a computer-implemented method executed under the control of one or more processing devices is provided. The method includes receiving measurements of a first portion of particles associated with an experiment from a particle analyzer. The method includes converting the measurements into initial transformed measurements using a non-parametric transformation. The method includes generating a parametric model to receive the measurements of the first portion of particles associated with the experiment as input and to generate the initial transformed measurements as output. The method also includes configuring the particle analyzer. The particle analyzer is configured to convert measurements of particles included in a second portion of the particles based at least in part on the parametric model, and to classify the particles based at least in part on the measurements and a sorting criterion.

[0017] Some embodiments of the method include receiving gate information that identifies a range of measurement values ​​in a non-parametric space for classifying the particles, wherein the sorting criteria comprises the gate information.

[0018] Some embodiments of the method include transmitting the parameter model to the particle analyzer and causing the particle analyzer to configure sorting circuitry based at least in part on the parameter model and the sorting criteria. For example, the sorting circuitry can be implemented in the form of a field programmable gate array.

[0019] The measurements received from the particle analyzer may include measurements of light emitted in the form of fluorescence by the first portion of particles. The light emitted in the form of fluorescence by the first portion of particles may include light emitted in the form of fluorescence by antibodies bound to the first portion of particles.

[0020] Some embodiments of the method can include selecting a measurement value and an initial transformed measurement value for a particle, and generating a composite measurement value pair based at least in part on: (i) the measurement value of the particle, (ii) the initial transformed measurement value of the particle; and (iii) a noise value, wherein the composite measurement value pair includes the composite measurement value and the composite transformed measurement value. The parametric model can further receive the composite measurement value as input and generate the composite transformed measurement value as output. In some embodiments, the method can also include receiving gate information that identifies a range of measurement values ​​for classifying the particle. For example, the particle can be selected based at least in part on the measurement value corresponding to the range of measurement values.

[0021] Some embodiments of the method may include receiving a hardware identifier from the particle analyzer, the hardware identifier indicating sorting circuitry implemented in the particle analyzer, and determining an architecture of the parameter model based at least in part on the hardware identifier, wherein the architecture of the parameter model is applicable to the sorting circuitry. In such embodiments, generating the parameter model may include generating the parameter model consistent with the architecture. For example, the parameter model may include a neural network, wherein the architecture identifies the number of layers of the neural network and the number of nodes per layer.

[0022] In an embodiment where the parametric model comprises a neural network, generating the parametric model may include assigning initial weights to connections between a first node in a first layer of the parametric model and a second node in a second layer of the parametric model; generating an output applicable to a measurement of a particle; comparing the output to the transformed measurement of the particle; and adjusting the initial weights based at least in part on a difference between the output and the transformed measurement, wherein a subsequent difference for the particle is less than the difference after adjusting the initial weights.

[0023] In another innovative aspect, a system is provided that includes one or more processing devices and a computer-readable storage medium containing instructions. The instructions, when executed by the one or more processing devices, cause the system to receive measurements of a first portion of particles associated with an experiment from a particle analyzer; convert the measurements into initial transformed measurements using a non-parametric transformation; generate a parametric model that receives as input the raw measurements of the first portion of particles associated with the experiment and generates as output the initial transformed measurements; and configure the particle analyzer. The particle analyzer is configured to convert measurements of particles included in a second portion of the particles based at least in part on the parametric model; generate a control signal to adjust an operating state of an analysis device included in the particle analyzer; and transmit the control signal to the analysis device to achieve the operating state.

[0024] In some embodiments, the analysis device may include sorting electronics communicatively coupled to the deflection plates. The particle analyzer may be further configured to identify a target container suitable for the particles based at least in part on the converted measurements corresponding to sorting criteria associated with the target container. Adjusting the operating state may include applying an electrical charge through the deflection plates to direct the particles into the target container. In some embodiments, the analysis device may include a fluidics system, and adjusting the operating state includes adjusting the pressure applied during the experiment. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A functional block diagram of an example of a transformation control system for analyzing and displaying biological events is shown.

[0026] Figure 2A is a schematic diagram of a particle sorter system according to one embodiment described herein.

[0027] Figure 2B is a schematic diagram of a particle sorting system 200 according to an embodiment of the present invention.

[0028] Figure 3 A functional block diagram of a particle analysis system for computation-based sample analysis and particle characterization is shown.

[0029] Figure 4 is a simplified diagram illustrating an example system for dynamically transforming event data.

[0030] Figure 5 is a simplified diagram showing an example system that uses parametric machine learning transformations to dynamically transform event data.

[0031] Figure 6is a process flow diagram depicting an example of a parametric machine learning transformation method applicable to multidimensional event data.

[0032] Figure 7 is a process flow diagram depicting an example of a method for sample sorting using parametric machine learning transformations. DETAILED DESCRIPTION

[0033] Before describing the present invention in more detail, it should be understood that the present invention is not limited to the specific embodiments described, as variations in actual implementation are inevitable. It should also be understood that the terminology used herein is for the purpose of describing specific embodiments only and is not intended to limit the present invention, which will be defined solely by the appended claims.

[0034] Where a numerical range is provided, it is understood that each intervening value between the upper and lower limits of the range and any other specified value or intervening value within the range is encompassed within the scope of the invention. Unless the context clearly dictates otherwise, each intervening value should be down to one-tenth of the unit of the lower limit. The upper and lower limits of these smaller ranges may independently be included in the smaller ranges and are also encompassed within the invention, subject to the requirements of any specifically excluded limits within the range. Where the range includes one or both limits, ranges excluding either or both of those included limits are also encompassed within the invention.

[0035] Certain ranges set forth herein are preceded by the term "about." The term "about" is used herein to provide literal support for the precise number that follows, as well as for a number that is close to or approximately the number that follows the term. In determining whether a number is close to or approximately a specifically recited number, the close or approximate unrecited number may be a number that is substantially equivalent to the specifically recited number in the context in which it appears.

[0036] 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 to which the invention belongs. Although methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, representative illustrative methods and materials are described below.

[0037] All publications and patents cited in this specification are incorporated herein by reference as if each individual publication or patent was specifically and individually indicated to be incorporated by reference and are incorporated herein for the purpose of disclosing and describing the methods and / or materials in connection with the cited publications. The citation of any publication is for its disclosure prior to the filing date and should not be construed as an admission that the present invention is not entitled to antedate such publication by virtue of prior invention. Furthermore, the dates of publication provided may be different from the actual publication dates, which may need to be independently confirmed.

[0038] It is noted that, as used herein and in the appended claims, the singular forms "a," "an," and "the" include plural referents unless the context clearly dictates otherwise. It is also noted that claims may be drafted to exclude any optional element. Thus, this statement is intended to serve as antecedent basis for use of terminology such as "solely," "only," and the like in connection with reciting claim elements or for use of a "negative" limitation.

[0039] After reading this disclosure, it will be apparent to those skilled in the art that each individual embodiment described and listed in this patent has layered components and features that can be quickly decomposed or combined with the features of any of the other embodiments without departing from the scope and spirit of the invention. Any recited method can be implemented in the order of events recited or in any other order that is logically possible.

[0040] Although the apparatus and method have been or will be described in a functional manner for grammatical fluency, it should be clearly understood that, except as expressly provided in 35 U.S.C., the claims shall not be construed as necessarily limiting the "means" or "steps" but shall be given the full meaning and scope of equivalents in accordance with the doctrine of equivalents, and when the claims are expressly drafted in accordance with 35 U.S.C. § 112, the claims shall be given full compliance with the statutory equivalents in 35 U.S.C. § 112.

[0041] Non-parametric transformations, such as t-distributed stochastic neighbor embedding (tSNE), can be used to analyze multi-parametric data, such as data from flow cytometry or other particle analysis systems and methods. The present invention can include these transformations for dimensionality reduction and subpopulation identification (e.g., gating). Essentially, non-parametric transformations cannot transform new observations without training the new transformation on the entire data set (including the new observations). This presents a significant challenge for computationally based particle analysis and sample processing based on such analysis (e.g., sorting, instrument control, etc.). The functionality utilizes neural networks (e.g., artificial neural networks or other computational models trained to approximate the non-parametric transformation) to parameterize non-parametric transformations, thereby allowing small training data sets to be transformed using non-parametric techniques while allowing accurate parametric models to be generated using the small training data set so that other events can be evaluated in a manner consistent with the initial event.

[0042] In some embodiments, a small training data set is transformed using a non-parametric transformation such as tSNE (e.g., tens of thousands of examples). The small training data set can be used to synthesize a large training data set. For example, transformed / untransformed pairs are randomly sampled from the small training data set, and random noise is added to the transformed and untransformed data. The random sampling can be arbitrarily weighted, for example, by density, target population, or uniform weight. The added noise can be based on a probability distribution, including uniform, Gaussian, or Poisson distributions.

[0043] The resulting large training data set is used to train a neural network to perform the transformations produced by the non-parametric transformations (which are performed on the small training data set). The neural network can have a variety of architectures, with any number of layers, any number of neurons in a layer, and any activation function between connected neurons. However, in order to facilitate the conversion of the neural network to the configuration of the sorting electronics included in the analytical instrument, the network training can be restricted to an architecture that can be converted to the target analytical instrument. For example, the neural network can consist of an entry layer (whose number of nodes is equal to the dimension of the untransformed data), four fully connected rectified linear unit (re-lu) activation layers (each with 30 nodes), and an output layer (with two nodes, linear activation). Training can be performed using a variety of error functions (e.g., mean squared error, Kullback-Leibler divergence) and optimization procedures (e.g., Adadelta, RMSProp).

[0044] After training, the neural network approximates the transformation originally used to transform the small training dataset. Thus, the neural network represents the transformation between the space occupied by the original, untransformed data and the space occupied by the transformed data. Thus, the neural network can transform new observations by feeding new data into the input layer and collecting the resulting transformed data in the output layer. The output data can then be used in downstream applications, such as cell sorting decisions.

[0045] The functionality described herein is a significant improvement over existing particle analysis systems, such as those that include non-parametric transformations. Non-parametric transformations are commonly used to analyze multi-parametric data, such as flow cytometry data. For example, tSNE is used to perform dimensionality reduction on flow cytometry data to visualize cell subpopulations. A fundamental limitation of non-parametric transformations is that, once trained, they cannot be used to transform new observations. Therefore, introducing new data requires re-performing the entire transformation, which often results in a completely different transformation. Consequently, non-parametric transformations have limited utility in areas where new observations need to be analyzed, such as FACS. The functionality described herein overcomes these limitations through non-parametric transformations.

[0046] Other ways of parameterizing nonparametric transformations rely on large training sets (e.g., about 100,000 or more data points). The technology described herein utilizes small training sets (which can be used to synthesize representative data sets of sufficient size) to train machine learning models that produce accurate results using parameterized variations. This saves time and materials for users and is more compatible with the sample sizes common when evaluating biological samples. For example, a typical flow cytometry setup experiment typically includes tens of thousands of data points. The number of these data points may be too small to train an accurate neural network. Requiring users to collect hundreds of thousands to millions of data points is time-consuming and requires a large number of samples, which is usually expensive and difficult to obtain.

[0047] Other approaches to parameterizing non-parametric transformations rely on large, complex neural networks with many layers and nodes. These large networks are computationally expensive to evaluate and therefore cannot be implemented within sufficiently short computational times for low-latency applications, such as cell sorting on sorting electronics with limited processing resources. The technique described here utilizes a relatively small number of layers and nodes, allowing for low-latency implementation on resource-efficient sorting electronics such as field-programmable gate arrays (FPGAs).

[0048] In the case of flow sorting, the functionality allows for the collection of a small dataset similar to the one currently used to establish the sorting gating strategy. Non-parametric transformations can then be performed on the small dataset. For example, tSNE can be performed to reduce the dimensionality of the data. The system can then receive gates specified using the tSNE transformed data. The pre-tSNE data and the post-tSNE data are then used to train a neural network that learns the transformation between the original data space and the reduced dimensionality space. The trained neural network can then be programmed into an FPGA and used to sort cells (or other particles) for actual sorting experiments. In previous implementations, sorting could not be performed based on data transformed using non-parametric transformations.

[0049] In some embodiments, the transformation model includes a dynamic algorithm, such as a machine learning algorithm. In accordance with conventional meanings, the term "machine learning" as used herein refers to adjustments to programming by computational methods that directly utilize data to determine and implement information without relying on predetermined equations as models. In certain embodiments, machine learning includes a learning algorithm that looks for patterns in data signals (e.g., from a variety of particles in a flow cytometry sample). In these embodiments, the learning algorithm is configured to generate better, more accurate decisions and predictions based on the number of data signals (i.e., as the number of characterizing particles from the sample increases, the learning algorithm becomes more robust). Purpose machine learning schemes may include, but are not limited to, artificial neural networks, decision tree learning, decision tree predictive modeling, support vector machines, Bayesian networks, dynamic Bayesian networks, genetic algorithms, and other machine learning schemes.

[0050] A neural network or neural network model can be conceptualized as a network of nodes. The nodes can be organized into layers, where the first layer is an input layer where data flows in. The neural network can also include an output layer where transformed data flows out. Each individual node can have multiple inputs and a single output (e.g., an input layer node has only a single input). The output of a node represents a linear combination of the inputs. In other words, the inputs can be multiplied by a relevant constant. The product can be accumulated along a path of nodes with a constant bias. The constant bias or "bias" can represent another degree of freedom that can be adjusted during the training process. For example, for a neural network model based on Re-LU, the constant bias can be a threshold because it has the ability to reduce the node value to below zero, causing the activation function to output zero.

[0051] The resulting value is evaluated using an activation function and used as the output of the node. A node in a given layer of the neural network is connected to every node in an adjacent layer. Neural networks can be trained by comparing the desired output of the network with the actual output of the network using a gradient descent algorithm and an error function to minimize the error of the network. The weights of one or more nodes can be adjusted to model the desired outcome produced by the network.

[0052] Typically, neural networks require a large amount of data for training. In this article, a small data set can be used for training by bootstrapping (generating synthetic data based on the small data set). In accordance with the conventional meaning, the term "bootstrapping" as used herein refers to generating one or more additional data sets with a larger amount of data than the original data set, for example, generating a data set with 10 data points or more (e.g., 100 data points or more, 1000 data points or more, 100,000 data points or more, and including 1,000,000 data points or more) more than the original data set. In some cases, the data points included in the generated data set are increased by 2 times or more, for example, 3 times or more, 5 times or more, 10 times or more, 25 times or more, 50 times or more, 100 times or more, and including 1,000 times or more, compared to the original data set. In certain embodiments, bootstrapping includes random permutation sampling. In certain embodiments, bootstrapping includes estimating the characteristics of the estimation function (e.g., by variance) by measuring the characteristics when sampling from an approximate distribution (e.g., an empirical distribution function). Bootstrapping according to certain embodiments may include using data from the flow cytometer in combination with noise components generated by the flow cytometer, such as vibrations from the laser assembly or detection system and noise from the photodetector.

[0053] Careful selection of the neural network architecture to minimize the number of nodes per layer and the number of layers allows for implementation of the neural network in computing hardware (e.g., FPGAs). Additionally, selecting activation functions that are compatible with FPGAs can further facilitate hardware implementation of the network. Hardware implementation is crucial for data transformations with sufficiently low latency for cell sorting applications.

[0054] The terms used herein and specifically set forth below have the following definitions. Unless otherwise defined herein, all terms used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this invention belongs.

[0055] As used herein, "system," "apparatus," "device," and "equipment" generally encompass hardware (e.g., mechanical and electronic hardware) and, in some embodiments, related software (e.g., a dedicated computer program for graphics control) components.

[0056] As used herein, an "event" or "event data" generally refers to a data packet measured from a single particle (e.g., a cell or synthetic particle). Typically, the data measured from a single particle includes a number of parameters, including one or more light scattering parameters, and at least one parameter or feature derived from fluorescence detected from the particle, such as the fluorescence intensity. Thus, each event is represented as a measurement value and a feature vector, where each measured parameter or feature corresponds to a dimension of the data space. In some embodiments, the data measured from a single particle includes images, electrical data, time data, or acoustic data. An event can be associated with an experiment, an assay, or a sample source (which can be identified by the measurement data).

[0057] As used herein, a "population" or "subpopulation" of particles (e.g., cells or other particles) generally refers to a group of particles having a characteristic (e.g., optical, impedance, or time characteristic) with respect to one or more measured parameters that causes the measured parameter data to form clusters in the data space. Thus, a population is identified as a cluster in the data. In contrast, each data cluster is typically interpreted as corresponding to a particular type of cell or particle population, although clusters corresponding to noise or background are also typically observed. Clusters can be defined in a subset of the dimensions (e.g., a subset with respect to the measured parameters) that correspond to populations that differ only in a subset of the measured parameters or characteristics (extracted from the cell or particle measurements).

[0058] As used herein, "gate" generally refers to the boundaries of a classifier that identifies a subset of data of interest. In cytometry, a gate may be associated with a specific set of events of interest. As used herein, "gating" generally refers to the process of classifying a given dataset using a defined gate, where the gate can be one or more regions of interest combined with Boolean logic.

[0059] As used herein, an "event" generally refers to an assembly of data measured from a single particle (e.g., a cell or synthetic particle). Typically, the data measured from a single particle includes a number of parameters or features, including one or more light scattering parameters or features and at least one other parameter or feature derived from the measured fluorescence. Thus, each event is represented as a vector of parameter and feature measurements, where each measured parameter or feature corresponds to a dimension of the data space.

[0060] Aspects of the present invention also include systems having a light detection system to characterize particles in a sample within a fluid medium (e.g., cells in a biological sample). Systems according to certain embodiments include: a light source configured to irradiate a sample (having particles) in a fluid medium; a light detection system comprising a photodetector configured to generate a data signal using light detected from particles within the fluid medium; and a processor having a memory operably coupled thereto, wherein the memory has instructions stored thereon that, when executed by the processor, cause the processor to perform the following operations: apply a nonparametric transform to a first set of data signals from the light detection system to generate a first transformed data set; generate a second transformed data set using the first transformed data set; generate a transformation model using the second transformed data set; and apply the transformation model to a second set of data signals from the light detection system.

[0061] The target system includes a light source configured to irradiate a sample in a fluid medium. In an embodiment, the light source can be any suitable broadband or narrowband light source. Depending on the components in the sample (e.g., cells, microbeads, non-cellular particles, etc.), the light source can be configured to emit light of different wavelengths in the range of 200nm to 1500nm, for example, 250nm to 1250, 300nm to 1000nm, 350nm to 900nm, including 400nm to 800nm. For example, the light source may include a broadband light source that emits light with a wavelength range of 200nm to 900nm. In other cases, the light source includes a narrowband light source that emits light with a wavelength range of 200nm to 900nm. For example, the light source can be a narrowband LED (1nm–25nm) that emits light with a wavelength range of 200nm to 900nm.

[0062] In some embodiments, the light source is a laser. The target laser may include a pulsed laser or a continuous wave laser. For example, the laser may be a gas laser, such as a helium-neon laser, an argon laser, a krypton laser, a xenon laser, a nitrogen laser, a CO2 laser, a CO laser, an argon fluoride (ArF) excimer laser, a krypton fluoride (KrF) excimer laser, a xenon chloride (XeCl) excimer laser, or a xenon fluoride (XeF) excimer laser, or a combination thereof; or a metal vapor laser, such as a helium cadmium (HeCd) laser, a helium mercury (HeHg) laser, a helium selenium ( or a solid-state laser, for example, a ruby ​​laser, an Nd:YAG laser, an NdCrYAG laser, an Er:YAG laser, an Nd:YLF laser, an Nd:YVO4 laser, an Nd:YCa4O(BO3)3 laser, an Nd:YCOB laser, a titanium sapphire laser, a thulim YAG laser, a YAG ytterbium laser, an ytterbium trioxide laser or a cerium-doped laser, and a combination thereof; or a semiconductor diode laser, an optically pumped semiconductor laser (OPSL) or a frequency doubled or tripled implementation of any of the above lasers.

[0063] In other embodiments, the light source is a non-laser light source, such as a lamp (including but not limited to a halogen lamp, a deuterium arc lamp, a xenon arc lamp), a light-emitting diode (e.g., a broadband LED with a continuous spectrum, a superluminescent diode, a semiconductor light-emitting diode, a broadband LED white light source, or an integrated light source of multiple LEDs). In some cases, the non-laser light source is a stable fiber-coupled broadband light source, a white light source, or other light sources, or any combination thereof.

[0064] In some embodiments, the light source is a beam generator configured to generate two or more beams of frequency-shifted light. In some cases, the beam generator includes a laser and a radio frequency generator, and the radio frequency generator is configured to apply a radio frequency drive signal to the acousto-optic device to generate two or more angle-deflected laser beams. In these embodiments, the laser can be a pulsed laser or a continuous wave laser. For example, the laser in the target beam generator can be a gas laser, such as a helium-neon laser, an argon laser, a krypton laser, a xenon laser, a nitrogen laser, a CO2 laser, a CO laser, an argon fluoride (ArF) excimer laser, a krypton fluoride (KrF) excimer laser, a xenon chloride (XeCl) excimer laser or a xenon fluoride (XeF) excimer laser or a combination thereof; or a metal vapor laser, such as a helium cadmium (HeCd) laser, a helium mercury (HeHg) laser, Helium selenide (HeSe) laser, helium silver (HeAg) laser, strontium laser, neon copper (NeCu) laser, copper laser or gold laser and combinations thereof; or a solid-state laser, for example, a ruby ​​laser, Nd:YAG laser, NdCrYAG laser, Er:YAG laser, Nd:YLF laser, Nd:YVO4 laser, Nd:YCa4O(BO3)3 laser, Nd:YCOB laser, titanium sapphire laser, thulim YAG laser, Ytterbium YAG laser, ytterbium oxide laser or cerium-doped laser and combinations thereof.

[0065] The acousto-optic device can be any suitable acousto-optic solution configured to frequency-shift a laser using an applied acoustic wave. In some embodiments, the acousto-optic device is an acousto-optic deflector. The acousto-optic device in the target system is configured to generate an angularly deflected laser beam using light from a laser and an applied RF drive signal. The RF drive signal can be applied to the acousto-optic device using any suitable RF drive signal source, such as a direct digital synthesizer (DDS), an arbitrary waveform generator (AWG), or an electrical pulse generator.

[0066] In an embodiment, the controller is configured to apply a radio frequency drive signal to the acousto-optic device to generate a desired number of angularly deflected laser beams in the output laser beam, for example, configured to apply 3 or more radio frequency drive signals, 4 or more radio frequency drive signals, 5 or more radio frequency drive signals, 6 or more radio frequency drive signals, 7 or more radio frequency drive signals, 8 or more radio frequency drive signals, 9 or more radio frequency drive signals, 10 or more radio frequency drive signals, 15 or more radio frequency drive signals, 25 or more radio frequency drive signals, 50 or more radio frequency drive signals, including being configured to apply 100 or more radio frequency drive signals.

[0067] In some cases, in order to obtain the intensity distribution of the angularly deflected laser beam in the output laser beam, the controller is configured to apply RF drive signals with different amplitudes, for example, the amplitude ranges from about 0.001 V to about 500 V, about 0.005 V to about 400 V, about 0.01 V to about 300 V, about 0.05 V to about 200 V, about 0.1 V to about 100 V, about 0.5 V to about 75 V, about 1 V to 50 V, about 2 V to 40 V, 3 V to about 30 V, including about 5 V to about 25 V. In some embodiments, each applied RF drive signal has a frequency in the range of about 0.001 MHz to about 500 MHz, for example, about 0.005 MHz to about 400 MHz, about 0.01 MHz to about 300 MHz, about 0.05 MHz to about 200 MHz, about 0.1 MHz to about 100 MHz, about 0.5 MHz to about 90 MHz, about 1 MHz to about 75 MHz, about 2 MHz to about 70 MHz, about 3 MHz to about 65 MHz, about 4 MHz to about 60 MHz, including about 5 MHz to about 50 MHz.

[0068] In some embodiments, the controller includes a processor having a memory operably coupled thereto, such that the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam comprising an angularly deflected laser beam having a desired intensity distribution. For example, the memory may include instructions for generating two or more angularly deflected laser beams having the same intensity, e.g., 3 or more, 4 or more, 5 or more, 10 or more, 25 or more, or 50 or more, and may include instructions for generating 100 or more angularly deflected laser beams having the same intensity. In other embodiments, the memory may include instructions for generating two or more angularly deflected laser beams having different intensities, e.g., 3 or more, 4 or more, 5 or more, 10 or more, 25 or more, or 50 or more, and may include instructions for generating 100 or more angularly deflected laser beams having different intensities.

[0069] In some embodiments, the controller includes a processor having a memory operably coupled thereto such that the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam having an intensity along a transverse axis that gradually increases from an edge to a center thereof. In these cases, the intensity of the angularly deflected laser beam measured at the center of the output beam can be from 0.1% to about 99%, e.g., from 0.5% to about 95%, from 1% to about 90%, from about 2% to about 85%, from about 3% to about 80%, from about 4% to about 75%, from about 5% to about 70%, from about 6% to about 65%, from about 7% to about 60%, from about 8% to about 55%, including from about 10% to about 50% of the intensity of the angularly deflected laser beam measured at an edge of the output laser beam along the transverse axis. In other embodiments, the controller comprises a processor having a memory operatively coupled thereto such that the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam having an intensity that increases gradually from an edge to a center thereof along a transverse axis. In these cases, the intensity of the angularly deflected laser beam measured at an edge of the output beam can be from 0.1% to about 99% of the intensity of the angularly deflected laser beam measured at the center of the output laser beam along the transverse axis, e.g., from 0.5% to about 95%, from 1% to about 90%, from about 2% to about 85%, from about 3% to about 80%, from about 4% to about 75%, from about 5% to about 70%, from about 6% to about 65%, from about 7% to about 60%, from about 8% to about 55%, including from about 10% to about 50% of the intensity of the angularly deflected laser beam measured at the center of the output laser beam along the transverse axis. In other embodiments, the controller comprises a processor having a memory operably coupled thereto, such that the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam having an intensity distribution along the horizontal axis that conforms to a Gaussian distribution. In certain embodiments, the controller comprises a processor having a memory operably coupled thereto, such that the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate an output laser beam having an intensity distribution along the horizontal axis that conforms to a top-hat distribution.

[0070] In an embodiment, the target beam generator can be configured to generate spatially separated angularly deflected laser beams in the output laser beam. The spacing between the angularly deflected laser beams can be 0.001 according to the applied RF drive signal and the desired irradiation profile of the output laser beam. In some embodiments, the system is configured to generate an overlapping angular deflection laser beam overlap in the output laser beam, for example, an overlap with an adjacent angular deflection laser beam distributed along the horizontal axis of the output laser beam. The overlapping portion between adjacent angular deflection laser beams (e.g., beam spot overlap) can be an overlap of 0.001 μm or more, for example, an overlap of 0.005 μm or more, an overlap of 0.01 μm or more, an overlap of 0.05 μm or more, an overlap of 0.1 μm or more, an overlap of 0.5 μm or more, an overlap of 1 μm or more, an overlap of 5 μm or more, an overlap of 10 μm or more, including an overlap of 100 μm or more.

[0071] In some cases, a beam generator configured to generate two or more beams of frequency-shifted light includes a laser excitation module as described in U.S. Patent Nos. 9,423,353, 9,784,661, and 10,006,852; and U.S. Patent Publication Nos. 2017 / 0133857 and 2017 / 0350803, the contents of which are incorporated herein by reference.

[0072] In an embodiment, the system includes a light detection system having one or more of a bright field photodetector, a light scattering (forward light scattering, side light scattering) detector, and a fluorescence detector for detecting and measuring light from the sample. The target bright field, light scattering, and fluorescence detectors may include, but are not limited to, optical sensors, such as active pixel sensors (APS), avalanche photodiodes, image sensors, charge coupled devices (CCDs), intensified charge coupled devices (ICCDs), light emitting diodes, photon counters, bolometers, pyroelectric detectors, photoresistors, photovoltaic cells, photodiodes, photomultiplier tubes, phototransistors, quantum dot photoconductors or photodiodes, and combinations thereof, and other photodetectors. In certain embodiments, a charge coupled device (CCD), a semiconductor charge coupled device (CCD), an active pixel sensor (APS), a complementary metal oxide semiconductor (CMOS), an image sensor, or an N-type metal oxide semiconductor (NMOS) image sensor is used to measure light from the sample. In certain embodiments, the bright field photodetector includes an avalanche photodiode (APD). In some cases, the light scattering detector is an avalanche photodiode. In some cases, one or more of the fluorescence detectors is an avalanche photodiode.

[0073] In some embodiments, the optical detection system of interest includes a plurality of fluorescence detectors. In some cases, the optical detection system includes a plurality of solid-state detectors, such as photodiodes. In some cases, the optical detection system includes a fluorescence photodetector array, such as a photodiode array. In these embodiments, the photodetector array may include 4 or more photodetectors, for example, 10 or more photodetectors, 25 or more photodetectors, 50 or more photodetectors, 100 or more photodetectors, 250 or more photodetectors, 500 or more photodetectors, 750 or more photodetectors, including 1000 or more photodetectors. For example, the detector may be a photodiode array having 4 or more photodiodes, for example, 10 or more photodiodes, 25 or more photodiodes, 50 or more photodiodes, 100 or more photodiodes, 250 or more photodiodes, 500 or more photodiodes, 750 or more photodiodes, including 1000 or more photodiodes.

[0074] The photodetectors may be arranged in any geometric configuration as desired, wherein the desired configurations include, but are not limited to, square, rectangular, trapezoidal, triangular, hexagonal, heptagonal, octagonal, nonagonal, decagonal, dodecagonal, circular, elliptical, and irregular pattern configurations. The photodetectors in the photodetector array may be oriented at an angle to another photodetector (as described in the XZ plane) in the range of 10° to 180°, for example, 15° to 170°, 20° to 160°, 25° to 150°, 30° to 120°, including 45° to 90°. The photodetector array may be of any suitable shape, and may be a shape bounded by straight lines (e.g., square, rectangular, trapezoidal, triangular, hexagonal, etc.), a shape bounded by curves (e.g., circular, elliptical), and an irregular shape (e.g., a parabolic bottom connected to a flat top). In some embodiments, the photodetector array has a rectangular active surface.

[0075] The active surface width of each photodetector (e.g., photodiode) in the array ranges from 5 μm to 250 μm, for example, from 10 μm to 225 μm, from 15 μm to 200 μm, from 20 μm to 175 μm, from 25 μm to 150 μm, from 30 μm to 125 μm, including from 50 μm to 100 μm; and the length ranges from 5 μm to 250 μm, for example, from 10 μm to 225 μm, from 15 μm to 200 μm, from 20 μm to 175 μm, from 25 μm to 150 μm, from 30 μm to 125 μm, including from 50 μm to 100 μm. μm, 25 μm to 150 μm, 30 μm to 125 μm, including 50 μm to 100 μm; wherein the surface area of ​​each photodetector (e.g., photodiode) in the array ranges from 25 μm2 to 10000 μm2, for example, 50 μm2 to 9000 μm2, 75 μm2 to 8000 μm2, 100 μm2 to 7000 μm2, 150 μm2 to 6000 μm2, including 200 μm2 to 5000 μm2.

[0076] The size of the photodetector array can vary depending on the amount and intensity of light, the number of photodetectors, and the desired sensitivity, and its length can range from 0.01 mm to 100 mm, for example, 0.05 mm to 90 mm, 0.1 mm to 80 mm, 0.5 mm to 70 mm, 1 mm to 60 mm, 2 mm to 50 mm, 3 mm to 40 mm, 4 mm to 30 mm, including 5 mm to 25 mm. The width of the photodetector array can also vary, ranging from 0.01 mm to 100 mm, for example, 0.05 mm to 90 mm, 0.1 mm to 80 mm, 0.5 mm to 70 mm, 1 mm to 60 mm, 2 mm to 50 mm, 3 mm to 40 mm, 4 mm to 30 mm, including 5 mm to 25 mm. Thus, the surface area of ​​the active surface of the photodetector array ranges from 0.1 mm2 to 10000 mm2, such as 0.5 mm2 to 5000 mm2, 1 mm2 to 1000 mm2, 5 mm2 to 500 mm2, including 10 mm2 to 100 mm2.

[0077] The photodetector of interest is configured to measure light collected at one or more wavelengths, for example, 2 or more wavelengths, 5 or more different wavelengths, 10 or more different wavelengths, 25 or more different wavelengths, 50 or more different wavelengths, 100 or more different wavelengths, 200 or more different wavelengths, 300 or more different wavelengths, including measuring light emitted by a sample in a fluid medium at 400 or more different wavelengths.

[0078] In some embodiments, the light detection system includes a brightfield photodetector configured to generate a brightfield data signal. The brightfield photodetector may be configured to detect light from the sample at one or more wavelengths, for example, 5 or more different wavelengths, 10 or more different wavelengths, 25 or more different wavelengths, 50 or more different wavelengths, 100 or more different wavelengths, 200 or more different wavelengths, 300 or more different wavelengths, including detecting light at 400 or more different wavelengths. The brightfield photodetector may be configured to detect light at one or more wavelengths in the wavelength range of 200nm–1200nm. In some cases, the method includes detecting light from the sample with a brightfield photodetector within a certain wavelength range, for example, 200nm to 1200nm, 300nm to 1100, 400nm to 1000nm, 500nm to 900nm, including 600nm to 800nm.

[0079] In some embodiments, the brightfield photodetector in the target light detection system is configured to generate one or more brightfield data signals responsive to the detected light, such as 2 or more, 3 or more, 4 or more, 5 or more, and including 10 or more brightfield data signals responsive to the detected light. Where the brightfield photodetector is configured to detect light at multiple wavelengths of light (e.g., 400 nm to 800 nm), the method may, in some cases, include generating one or more brightfield data signals responsive to each wavelength of the detected light. In other cases, a single brightfield data signal is generated in response to light detected by the brightfield photodetector across the entire wavelength range.

[0080] In some embodiments, the light detection system includes a light scattering photodetector configured to generate a light scattering data signal. The light scattering photodetector may be configured to detect light from the sample at one or more wavelengths, for example, 5 or more different wavelengths, 10 or more different wavelengths, 25 or more different wavelengths, 50 or more different wavelengths, 100 or more different wavelengths, 200 or more different wavelengths, 300 or more different wavelengths, including detecting light at 400 or more different wavelengths. The light scattering photodetector may be configured to detect light at one or more wavelengths in the wavelength range of 200nm–1200nm. In some cases, the method includes detecting light from the sample with a light scattering photodetector within a certain wavelength range, for example, 200nm to 1200nm, 300nm to 1100, 400nm to 1000nm, 500nm to 900nm, including 600nm to 800nm.

[0081] In certain embodiments, the light scattering photodetector in the target light detection system is configured to generate one or more light scattering data signals in response to the detected light, such as 2 or more, 3 or more, 4 or more, 5 or more, and including 10 or more light scattering data signals in response to the detected light. Where the light scattering photodetector is configured to detect light at multiple wavelengths of light (e.g., 400 nm to 800 nm), the method may, in some cases, include generating one or more light scattering data signals in response to each wavelength of the detected light. In other cases, a single light scattering data signal is generated in response to light detected by the light scattering photodetector across the entire wavelength range.

[0082] The light detection system includes one or more bright field, light scattering or fluorescence detectors, for example, 2 or more, 3 or more, 4 or more, 5 or more, 6 or more, 7 or more, 8 or more, 9 or more, 10 or more, 15 or more, including 25 or more detectors. In an embodiment, each of the detectors is configured to generate a data signal. Each detector can be used to detect light from the sample independently at one or more wavelengths in the wavelength range of 200nm–1200nm. In some cases, one or more detectors are configured to detect light from the sample within a certain wavelength range, for example, 200nm to 1200nm, 300nm to 1100, 400nm to 1000nm, 500nm to 900nm, including 600nm to 800nm. In other cases, one or more detectors are configured to detect light at one or more specific wavelengths. For example, the fluorescence can be detected at one or more wavelengths of 450 nm, 518 nm, 519 nm, 561 nm, 578 nm, 605 nm, 607 nm, 625 nm, 650 nm, 660 nm, 667 nm, 670 nm, 668 nm, 695 nm, 710 nm, 723 nm, 780 nm, 785 nm, 647 nm, 617 nm, and any combination thereof, depending on the number of different detectors in the target light detection system. In some embodiments, one or more detectors are configured to detect wavelengths of light corresponding to peak fluorescence wavelengths of certain fluorophores in the sample.

[0083] The light detection system is configured to measure light continuously or at discrete time intervals. In some cases, the detectors in the light detection system are configured to measure the collected light continuously. In other cases, the light detection system is configured to measure at discrete time intervals, for example, at intervals of 0.001 milliseconds, 0.01 milliseconds, 0.1 milliseconds, 1 millisecond, 10 milliseconds, 100 milliseconds, including 1000 milliseconds, or some other time interval.

[0084] In some embodiments, the system is configured to generate frequency-encoded fluorescence data by irradiating a sample (having particles) in a fluid medium. In some embodiments, the light source includes a light generator assembly that generates multiple angularly deflected laser beams, each laser beam having an intensity based on the amplitude of an applied radio frequency drive signal (e.g., from a direct digital synthesizer coupled to an acousto-optic device). For example, the target system may include a light generator assembly that generates two or more angularly deflected laser beams, e.g., three or more, four or more, five or more, six or more, seven or more, eight or more, nine or more, ten or more, and including twenty-five or more angularly deflected laser beams. In embodiments, each of the angularly deflected laser beams has a different frequency, the frequency being offset by a predetermined radio frequency relative to the frequency of the input laser beam.

[0085] According to certain embodiments, the target system is configured to generate angularly deflected laser beams that are spatially offset from each other. Depending on the applied RF drive signal and the desired irradiation profile of the output laser beam, the target system can be configured to generate angularly deflected laser beams, and the spacing between the laser beams can be 0.001 μm or more, for example, 0.005 μm or more, 0.01 μm or more, 0.05 μm or more, 0.1 μm or more, 0.5 μm or more, 1 μm or more, 5 μm or more, 10 μm or more, 100 μm or more, 500 μm or more, 1000 μm or more, including 5000 μm or more. In some embodiments, the angularly deflected laser beams overlap, for example, with adjacent angularly deflected laser beams distributed along the horizontal axis of the output laser beam. The overlapping portion between adjacent angularly deflected laser beams (e.g., beam spot overlap) can be an overlap of 0.001 μm or more, for example, an overlap of 0.005 μm or more, an overlap of 0.01 μm or more, an overlap of 0.05 μm or more, an overlap of 0.1 μm or more, an overlap of 0.5 μm or more, an overlap of 1 μm or more, an overlap of 5 μm or more, an overlap of 10 μm or more, including an overlap of 100 μm or more.

[0086] In some embodiments, a system includes a processor having a memory operably coupled thereto, wherein the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate frequency-encoded fluorescence data by calculating the difference between the optical frequencies of incident overlapping beamlets of light on the fluid medium. In one example, a system includes a processor having a memory operably coupled thereto, wherein the memory includes instructions stored thereon that, when executed by the processor, cause the processor to calculate a beat frequency at each position on a horizontal axis of the fluid medium. In these embodiments, the frequency-encoded fluorescence emitted by the particle is a beat frequency corresponding to the difference between the frequency of a local oscillator beam (fLO) and the frequency of a radio frequency offset beamlet. For example, the frequency-encoded fluorescence data includes the beat frequency of the fLO-fRF offset beamlet. In the case where the irradiation of the fluid medium includes a local oscillator that spans the width of the fluid medium (e.g., the entire horizontal axis), the frequency-encoded fluorescence data includes a beat frequency corresponding to the difference between the frequency of the local oscillator beam (fLO) and the frequency of each RF offset beamlet (f1, f2, f3, f4, f5, f6, etc.). In these embodiments, the frequency-encoded fluorescence data may include a plurality of beat frequencies, each beat frequency corresponding to a position on the horizontal axis of the fluid medium.

[0087] In some embodiments, the subject system includes a particle sorting assembly for sorting particles (e.g., cells) in the sample. In some cases, the particle sorting assembly is a particle sorting module, such as the particle sorting modules described in U.S. Patent Publication No. 2017 / 0299493, filed on March 28, 2017, and U.S. Provisional Patent Application No. 62 / 752,793, filed on October 30, 2018, the contents of which are incorporated herein by reference. In some embodiments, the particle sorting assembly includes one or more droplet deflectors, such as the droplet deflectors described in U.S. Patent Publication No. 2018 / 0095022, filed on June 14, 2017, the contents of which are incorporated herein by reference.

[0088] In some embodiments, the subject system is a flow cytometry system. Suitable flow cytometry systems can include, but are not limited to, those described in the following publications: Ormerod (ed.), Flow Cytometry: A Practical Approach, Oxford University Press (1997); Jaroszeski et al. (eds.), Flow Cytometry Protocols, Methods in Molecular Biology 91, Humana Press (1997); Practical Flow Cytometry, 3rd ed., Wiley-Liss (1995); Virgo et al. (2012), Annals of Clinical Biochemistry, January; 49(pt 1): 17-28; Linden et al., Seminars in Thrombosis and Hemostasis, October 2004; 30(5): 502-11; Alison et al., J. Pathol., December 2010; 222(4): 335-344; and Herbig et al. (2007) Crit Rev Ther Drug Carrier Syst. 24(3): 203-255; the contents of these publications are incorporated herein by reference. In some cases, the flow cytometry system of interest includes a BD Biosciences FACSCanto™II flow cytometer, a BD Accuri™ flow cytometer, a BD Biosciences FACSCelesta™ flow cytometer, a BD Biosciences FACSLyric™ flow cytometer, a BD Biosciences FACSVerse™ flow cytometer, a BD Biosciences FACSymphony™ flow cytometer, a BD Biosciences LSRFortessa™ flow cytometer, a BD Biosciences LSRFortess™ X-20 flow cytometer, and a BD Biosciences FACSCalibur™ cell sorter, a BD Biosciences FACSCount™ cell sorter, a BD Biosciences FACSLyric™ cell sorter, a BD Biosciences Via™ cell sorter, a BD Biosciences Influx TM Cell sorter, BD Biosciences Jazz TM Cell sorter, BD Biosciences Aria TM Cell sorter and BD Biosciences FACSMelody TM Cell sorter, etc.

[0089] In some embodiments, the target particle sorting system is a flow cytometry system, for example, a system described in the following U.S. Patent Nos.: 10,006,852; 9,952,076; 9,933,341; 9,784,661; 9,726,527; 9,453,789; 9,200,334; 9,097,640; 9,095,494; 9,092,034; 8,975,595; 8,753,5 The contents of these documents are incorporated herein by reference in their entirety.

[0090] In some cases, the subject system is a flow cytometry system configured to characterize and image particles in a fluid medium by fluorescence imaging using radiofrequency labeled emission (FIRE), such as the systems described in Diebold et al., Nature Photonics, Vol. 7(10); 806-810 (2013); U.S. Patent Nos. 9,423,353, 9,784,661, and 10,006,852; and U.S. Patent Publication Nos. 2017 / 0133857 and 2017 / 0350803, the contents of which are incorporated herein by reference.

[0091] Specific examples of various embodiments and systems in which the embodiments and systems may be implemented are described further below.

[0092] Figure 1 A functional block diagram of an example of a transformation control system for analyzing and displaying biological events is shown. The analysis controller 190 can be configured to implement various processes for controlling the graphical display of biological events.

[0093] The particle analyzer 102 can be configured to collect biological event data. For example, a flow cytometer can generate flow cytometry event data. The particle analyzer 102 can be configured to provide the biological event data to the analysis controller 190. A data communication channel can be included between the particle analyzer 102 and the analysis controller 190. The biological event data can be provided to the analysis controller 190 via the data communication channel.

[0094] Analysis controller 190 may be configured to receive biological event data from particle analyzer 102. The biological event data received from particle analyzer 102 may include flow cytometry event data. Analysis controller 190 may be configured to provide a graphical display including a first graph of the biological event data to display device 106. Analysis controller 190 may be further configured to render a region of interest as a gate around the plurality of biological event data displayed by display device 106, overlaid on the first graph. In some embodiments, the gate may be a logical combination of one or more graphical regions of interest drawn on a single parameter histogram or a bivariate graph.

[0095] Analysis controller 190 may further be configured to display the bio-event data on display device 106 inside the door in a manner different from other events in the bio-event data outside the door. For example, analysis controller 190 may be configured to display the bio-event data contained inside the door in a different color than the bio-event data contained outside the door. Display device 106 may be implemented in the form of a monitor, tablet computer, smartphone, or other electronic device configured to display a graphical interface.

[0096] The analysis controller 190 may be configured to receive a door selection signal identifying the door from a first input device. For example, the first input device may be implemented in the form of a mouse 110. The mouse 110 may send a door selection signal to the analysis controller 190 to determine the door to be displayed on the display device 106 or manipulated via the display device 106 (e.g., clicking on or in the desired door when the cursor is located there). In some embodiments, the first device may be implemented in the form of a keyboard 108 or other device for providing input signals to the analysis controller 190 (e.g., a touch screen, a stylus, an optical detector, or a voice recognition system). Some input devices may include multiple input functions. In such embodiments, the input functions may each be considered an input device. For example, Figure 1 As shown, the mouse 110 may include a right mouse button and a left mouse button, both of which may generate trigger events.

[0097] The triggering event may cause the analysis controller 190 to change the manner in which the data is displayed (actually displaying a portion of the data on the display device 106 ), or provide input for further processing, such as selecting a target population for particle sorting.

[0098] In some embodiments, analysis controller 190 can be configured to detect when mouse 110 initiates gate selection. Analysis controller 190 can further be configured to automatically modify the graph visualization to facilitate the gating process. The modification can be based on a specific distribution of bio-event data received by analysis controller 190.

[0099] Analysis controller 190 may be coupled to storage device 104. Storage device 104 may be configured to receive and store biological event data from analysis controller 190. Storage device 104 may also be configured to receive and store flow cytometry event data from analysis controller 190. Storage device 104 may also be configured to allow analysis controller 190 to retrieve biological event data, e.g., flow cytometry event data.

[0100] Display device 106 can be configured to receive display data from analysis controller 190. The display data can include graphs of biological event data and gates outlining portions of the graphs. Display device 106 can be further configured to change the displayed information based on input received from analysis controller 190 and input received from particle analyzer 102, storage device 104, keyboard 108, and / or mouse 110.

[0101] A common flow sorting technique (referred to as "electrostatic cell sorting") employs droplet sorting, in which a stream or moving column of liquid containing linearly separated particles is broken into droplets, and the droplets containing the particles of interest are charged and deflected into a collection tube by an electric field. Current droplet sorting systems are capable of forming droplets at a rate of 100,000 drops per second in a fluid medium passing through a nozzle having a diameter of less than 100 microns. Droplet sorting typically requires that the droplets be detached from the stream at a certain distance from the nozzle tip. The distance is typically on the order of a few millimeters from the nozzle tip and is stable for an undisturbed fluid medium and can be maintained by oscillating the nozzle tip at a predetermined frequency and an amplitude that keeps the detachment constant. For example, in some embodiments, the amplitude of a sinusoidal voltage pulse is adjusted at a given frequency to keep the detachment stable and constant.

[0102] Typically, linearly separated particles in the stream are characterized by their passage through an observation point located within a flow cell or cuvette, or directly below the nozzle tip. Once a particle has been determined to meet one or more of the required criteria, it is possible to predict when it will reach the droplet breakup point and separate from the stream as a droplet. Ideally, a brief charge is applied to the fluid medium before the droplets containing the selected particles separate from the stream, and then the droplets are immediately grounded after breakup. The droplets to be sorted remain charged upon separation from the fluid medium, while all other droplets remain uncharged. The charged droplets are laterally deflected from the downward trajectory of the other droplets by the electric field and are collected in a sample tube. The uncharged droplets fall directly into a drain tube.

[0103] Figure 2A is a schematic diagram of a particle sorter system according to one embodiment described herein. Figure 2AThe particle sorter system 250 shown includes deflection plates 252 and 254. A charge is applied via streamer wires in barbs 256. This creates a particle stream 260 for analysis. The particles can be illuminated with one or more light sources (e.g., lasers) to produce light scattering and generate fluorescence information. The particle information is detected by, for example, sorting electronics or other detection systems ( Figure 2A The deflector plates 252 and 254 can be independently controlled to attract or repel the charged droplets to direct the droplets to a destination collection container (e.g., one of 272, 274, 276, or 278). Figure 2A As shown, the deflector plates can be controlled to direct the particles along the first path 262 toward the container 274 or along the second path 268 toward the container 278. If the particles are not particles of interest (e.g., within a specified sorting range, not showing scattering or illumination information), the deflector plates can cause the particles to continue flowing along the flow path 264. Such uncharged droplets can be directed to a waste container, such as via an aspirator 270.

[0104] The sorting electronics may be included to begin collecting measurements, receive the fluorescence signal of a particle, and determine how to adjust the deflection plates to sort the particle. Figure 2A The exemplary embodiment shown in FIG. 1 includes a BD FACSAria commercially available from Becton, Dickinson and Company in San Jose, California. TM Serial flow cytometer.

[0105] Figure 2B is a schematic diagram of a particle sorter system 200 according to one embodiment described herein. In some embodiments, the particle sorter system 200 is a cell sorter system. Figure 2B As shown, a droplet formation sensor 202 (e.g., a piezoelectric oscillator) is coupled to a fluid conduit 201 (e.g., a nozzle). Within the fluid conduit 201, a sheath fluid 204 hydrodynamically focuses a sample fluid 206 into a moving liquid column 208 (e.g., a stream). Within the moving liquid column 208, particles (e.g., cells) are aligned to pass through a monitored region 210 (e.g., a laser stream intersection) irradiated by an irradiation source 212 (e.g., a laser). The droplet formation sensor 202 vibrates, causing the moving liquid column 208 to break into multiple droplets 209.

[0106] In operation, a detection station 214 (e.g., an event detector) determines when a particle of interest (or cell of interest) passes through the monitored area 210. The detection station 214 feeds a timing circuit 228, which in turn feeds a transient charging circuit 230. At the droplet breakup point, after notification of a timed droplet delay (Δt), a transient charge is applied to the moving liquid column 208 to charge the droplet of interest. The droplet of interest may comprise one / or more particles or cells to be sorted. The charged droplets may then be sorted by activating a deflection plate (not shown) to deflect the droplets into a container such as a collection tube or a multi-well sample plate, where a well may be associated with a particular droplet of interest. Figure 2B As shown, the droplets are collected in a drain container 238 .

[0107] A detection system 216 (e.g., a droplet boundary detector) is used to automatically determine the phase of the droplet drive signal as the particle of interest passes through the monitored area 210. An exemplary droplet boundary detector is described in U.S. Patent No. 7,679,039, which is incorporated herein by reference in its entirety. The detection system 216 allows the instrument to accurately calculate the position of each detected particle within the droplet. The detection system 216 can be fed with an amplitude signal 220 and / or a phase 218 signal, which in turn are fed (via amplifier 222) into an amplitude control circuit 226 and / or a frequency control circuit 224. The amplitude control circuit 226 and / or the frequency control circuit 224, in turn, control the droplet formation sensor 202. The amplitude control circuit 226 and / or the frequency control circuit 224 can be included in the control system.

[0108] In some embodiments, the sorting electronics (e.g., detection system 216, detection station 214, processor 240) can be coupled to a memory configured to store detected events and sorting decisions based thereon. The sorting decision can be included in the event data for the particle. In some embodiments, the detection system 216 and the detection station 214 can be implemented as a single detection unit or coupled in a communicative manner so that event measurements can be collected by one of the detection system 216 or the detection station 214 and provided to the non-collecting element.

[0109] In some embodiments, one or more of the components described for particle sorter system 200 may be used for particle analysis and characterization, regardless of whether the particles are physically sorted into collection containers. Similarly, one or more of the components described for particle analysis system 300 ( Figure 3 One or more of the components described below can also be used for particle analysis and characterization, regardless of whether the particles are physically sorted into collection containers. For example, one or more of the components described above can be used in the particle sorter system 200 or the particle analysis system 300 to group particles or display them in a tree that includes at least three of the groupings described herein.

[0110] Figure 3 A functional block diagram of a particle analysis system for computation-based sample analysis and particle characterization is shown. In some embodiments, particle analysis system 300 is a flow system. Figure 3 The particle analysis system 300 shown can be configured to perform, in whole or in part, the methods described herein, for example, Figure 5 The particle analysis system 300 includes a fluidics system 302. The fluidics system 302 may include or be coupled to a sample tube 310 and a moving liquid column within the sample tube, within which particles 330 (e.g., cells) in a sample move along a common sample path 320.

[0111] The particle analysis system 300 includes a detection system 304 configured to collect a signal from each particle as it passes through one or more detection stations along the common sample path. Detection stations 308 are generally referred to as monitored regions 340 of the common sample path. In some embodiments, detection can include detecting light or one or more other characteristics of the particles 330 as they pass through the monitored regions 340. Figure 3 , a detection station 308 is shown having a monitored area 340. Some embodiments of particle analysis system 300 may include multiple detection stations. In addition, some detection stations may monitor more than one area.

[0112] Each signal is assigned a signal value to form a data point for each particle. As described above, the data can be referred to as event data. The data point can be a multi-dimensional data point that includes values ​​for each measured characteristic of the particle. The detection system 304 is configured to collect a series of the data points over a first time interval.

[0113] The particle analysis system 300 also includes a control system 306. The control system 306 may include one or more processors, amplitude control circuit 226 and / or frequency control circuit 224 (e.g., Figure 2B ). As shown, control system 306 is operatively associated with fluidic system 302. Control system 306 is configured to generate a calculated signal frequency for at least a portion of the first time interval based on a Poisson distribution and a number of data points collected by detection system 804 during the first time interval. Control system 306 is further configured to generate an experimental signal frequency based on the number of data points during the portion of the first time interval. Furthermore, control system 306 can compare the experimental signal frequency to the calculated signal frequency or a predetermined signal frequency.

[0114] Figure 4is a simplified diagram illustrating an example system for dynamically transforming event data. System 400 includes a transform device 420. Transform device 420 includes an event data receiver 422. Event data receiver 422 can receive data from a particle analyzer (e.g., Figure 1 In some embodiments, the event data 402 may be generated by the particle analyzer by receiving the event data 402 from, for example, an analysis workstation. For example, a user may provide the event data 402 obtained from the particle analyzer to the event data receiver 422. The event data receiver 422 may include a transceiver for wireless communication or a port for connecting to a wired network such as an Ethernet local area network or a network such as via a universal serial bus or a serial bus. The connector connects to the port on the device.

[0115] The event data receiver 422 can provide at least a portion of the event data 402 to a transform selection circuit 424 included in the transform device 420. The transform selection circuit 424 can identify a transform that is applicable to the event data. The identification can include detecting a value in the event data, such as an identifier for the assay or experiment. The available transforms can be stored in a data memory 440 accessible to the transform selection circuit 424. As discussed above, the transform can be a parametric transform or a non-parametric transform. In some embodiments, the transform can be specified by the device providing the event data 402. For example, an analysis workstation can submit a message requesting that the event data 402 be processed. The message can include the desired transform (e.g., tSNE).

[0116] Transformation device 420 may include event data transformation circuitry 426 to process event data 402 according to the identified transformation. Event data transformation circuitry 426 may generate transformed event data 490. Transformed event data 490 may be transmitted from transformation device 420 to a destination device, such as a device that verifies event data 402, a storage location, or an addressable network service (e.g., a laboratory information system).

[0117] As mentioned above, when performing non-parametric transformations, consistent processing of event data received at different times is unpredictable due to the statistical nature of the transformation. Figure 4In the embodiment of the present invention, the conversion device 420 can provide suitable results when the measured value of event data 402 represents the completion of the experiment.But, for some analytical methods (for example, establishing sorting criteria), the user can make best guesses to the expected position of the target group, or obtain the measured value of a part of the sample, and then use it to specify the gate.In the first case, the guess may not be able to effectively find out the real target group, and often may need to obtain a larger sample size, which may be impractical for some experiments.In the situation of analyzing the initial part, if nonparametric transformation is used to transform the event data, then the gate of any gained all requires that the event in the future be transformed in the same mode as the initial part of the event.As mentioned above, nonparametric transformation is uncertain normally, and therefore the event data processed at two different times may provide two different results.

[0118] Figure 5 is a simplified diagram showing an example system that uses parametric machine learning transformations to dynamically transform event data. Figure 5 The transformation device 520 in includes the ability to parameterize non-parametric transformations using neural networks, thereby allowing small training data sets to be transformed using non-parametric techniques, while allowing accurate parametric models to be generated using the small training data sets to evaluate other events in a manner consistent with the initial event.

[0119] System 500 includes a transform device 520. Transform device 520 includes an event data receiver 522. Event data receiver 522 can receive data from a particle analyzer (e.g., Figure 1 In some embodiments, the event data 502 may be generated by the particle analyzer by receiving data from, for example, an analysis workstation. For example, a user may provide the event data 502 obtained from the particle analyzer to the event data receiver 522. The event data receiver 522 may include a transceiver for wireless communication or a network port for connecting to a wired network such as an Ethernet local area network.

[0120] Event data receiver 522 can activate switch 524 or other routing elements for further processing. Switch 524 can be connected in a branch line between transform selection circuit 540 and training circuit 530. Switch 524 can be activated based on: event data 502; information included in or with event data 502, such as a message requesting processing of event data 502; characteristics of the device providing event data 502; or other factors detectable by event data receiver 522 or switch 524.

[0121] When switch 524 activates a path to transition selection circuit 540, at least a portion of event data 502 is provided to transition selection circuit 540. Figure 4 In general, the transform selection circuit 540 can also identify a transform that is applicable to the event data. The identification can include detecting a value in the event data, such as an identifier for the assay or experiment. The available transforms can be stored in a data memory 540 accessible to the transform selection circuit 540. As discussed above, the transform can be a parametric transform or a non-parametric transform. In some embodiments, the transform can be specified by the device providing the event data 502. For example, an analysis workstation can submit a message requesting that the event data 502 be processed. The message can include the desired transform (e.g., tSNE).

[0122] The transformation device 520 may include event data transformation circuitry 542 to process the event data 502 according to the identified transformation. The event data transformation circuitry 542 may generate transformed event data 570. The transformed event data 570 may be transmitted from the transformation device 520 to a target device, such as a device that verifies the event data 502, a storage location, or an addressable network service (e.g., a laboratory information system).

[0123] In a second mode, switch 524 can be activated to route at least a portion of the event data 502 to a training circuit 530. Training circuit 530 can coordinate the generation of a parameter model for transforming the event data 502. Transformation device 520 can include a training event data generator 532 to generate a suitable training dataset for the machine learning model. The neural network can have a variety of architectures, with any number of layers, any number of neurons in a layer, and any activation function between connected neurons. However, to facilitate conversion of the neural network to the configuration of the sorting electronics included in the analytical instrument, network training can be restricted to architectures that can be converted to the target analytical instrument. For example, the neural network can consist of an input layer (with a number of nodes equal to the dimension of the untransformed data), four fully connected rectified linear unit (Re-LU) activation layers (each with 30 nodes), and an output layer (with two nodes and linear activation). Training can be performed using a variety of error functions (e.g., mean squared error, Kullback-Leibler divergence) and optimization procedures (e.g., Adadelta, RMSProp).

[0124] After training, the neural network approximates the transformation originally used to transform the small training dataset. Thus, the neural network represents the transformation between the space occupied by the original, untransformed data and the space occupied by the transformed data. Thus, the neural network can transform new observations by feeding the new data into the input layer and collecting the resulting transformed data in the output layer.

[0125] The generated event data can then be transmitted to the transformation device 520 for processing. The transformation of the training data can include a non-parametric transformation. The non-parametric transformation can be identified in the request message accompanying or associated with the generated event data. The generated event data can be processed via the event data receiver 522, the transformation selection circuit 540, and the event data transformation circuit 542 as described above to generate the transformed event data.

[0126] The transformed event data and the generated event data can then be received from the training circuit 530 to generate a transformation model. Generating the transformation model can include generating a neural network that approximates the transformation from the generated event data to the transformed event data. As discussed above, the construction of the model can be constrained based on factors such as the target particle analyzer or sorting electronics. This can facilitate the generation of a transformation model that can be applied to the target particle analyzer or sorting electronics. The particle analyzer or sorting electronics can be indicated using a hardware identifier that can be associated with the model architecture or constraints.

[0127] Transformation model 560 may be stored in data storage 514 and used for subsequent transformation of event data for the experiment or other experiments.

[0128] Figure 6 is a process flow diagram depicting an example of a parametric machine learning transformation method applicable to multi-dimensional event data. Figure 5 The transformation device 520 shown is implemented in whole or in part.

[0129] Multidimensional event data 602, such as particle analyzer data, can be provided. Multidimensional event data 602 can be processed using a non-parametric transformation 604 (e.g., tSNE) to generate transformed event data 606 representing the initial transformed measurement value of the event. Transformed event data 606 can be used to train a model through machine learning 608. The result of the learning can include a parameter transformation neural network model 610. The neural network model 610 can generate transformed data 614 for the multidimensional event data 602 or new multidimensional event data 612. Then, the transformed data 614 can be provided for other processing or control, such as gate selection 680, event sorting determination 682, or instrument adjustment 684 (e.g., pressure regulation of a flow cytometer). Gate selection 680 can include transmitting the transformed event data to a device to display one or more graphical user interfaces. A graphical user interface can include control elements for selecting a destination event and assigning a sorting destination (e.g., container, plate, hole, chamber, tube, etc.) to the event. The selection can include receiving gate information that identifies the measurement value range for classifying the particles. The gate can be specified using measurements in a transformed (eg, non-parametric) space. The gate information can form or be used to generate sorting criteria to configure the particle analyzer to sort particles suitable for the experiment.

[0130] Event sorting determination 682 may include transmitting the transformation and sorting criteria together to a particle analyzer. The particle analyzer (including sorting electronics or other analytical devices for processing particles during the experiment) may be configured to transform the raw event data and apply the specified sorting criteria. As discussed above, the transformation may be a parameterized neural network model that represents a non-parametric transformation for displaying events suitable for gate selection. Instrument adjustment 684 may include transmitting a control signal to the particle analyzer or an analytical device included therein to adjust its operating state. For example, the flow rate may be adjusted based on the distribution of the transformed event data compared to an ideal distribution. If the distribution deviates from the ideal distribution, the flow rate may be reduced. The amount of the reduction may be assessed using the size of the deviation from the ideal distribution. Other operating characteristics of the particle analyzer may be adjusted using the transformed event data (e.g., detector voltage or data scaling factor).

[0131] Figure 7 is a process flow diagram depicting an example of a method for sample sorting using a parametric machine learning transformation. Figure 1 Aspects of method 700 may be implemented in whole or in part using a management device (e.g. Figure 5 Method 700 illustrates how to process event data to generate a parametric transformation based on a non-parametric transformation of an initial set of events.

[0132] Method 700 begins at block 702. At block 710, initial event data is collected for a portion of a sample. The collection may include activating a particle analyzer to process the portion of the sample. Processing the sample may include measuring properties of the particles, such as graphical, electrical, temporal, or acoustic properties.

[0133] At block 720, the initial event data may be transmitted to a transformation device, such as Figure 5 Transformation device 520 is shown. The transmission may include transmitting a message requesting a transformation of the initial event data. The message may include an identifier applicable to the experiment, sample, sample source, or other unique information to identify the event data. The identifier may be used to select a transformation applicable to the event data. The identifier may be used to associate a parameter transformation generated using the event data with a subsequent transformation request.

[0134] At box 730, the initial event data is transformed using a non-parametric transformation. The transformation can be selected from a set of transformations based on, for example, the event data. For example, if the event data includes acoustic measurements, a tSNE transformation customized for acoustic values ​​can be used. In some embodiments, the transformation can be specified in a message requesting processing of the event data. For example, if a researcher requests a chart of the event data for gating, the request can include an indication of the type of transformation to be applied so as to graphically represent the dimensions of the event data for gating.

[0135] At block 740, the initial event data can be amplified to generate model training event data. The amplification can include generating training data based on the measured values ​​of the event. For example, the measured value can be used to generate a second synthetic event within a threshold distance of the measured value. The synthetic measured value can include a measurement pair, where one value is a synthetic untransformed measured value and the second value is a synthetic transformed measured value. Table 1 lists event data suitable for using a measurement value with a threshold of 0.001 (e.g., a synthetic measured value based on the actual measured value should be within 0.001 of the actual measured value).

[0136] Table 1

[0137] Event Id Is it synthetic? Measurement 1 no 2.30495 2 yes 2.30492 3 yes 2.30499 4 yes 2.30404

[0138] The threshold value may be determined based on the type of measurement. For example, an optical measurement (e.g., a fluorescence-based measurement) may have a greater tolerance than an image-based measurement. The threshold value may be specified in a configuration accessible to the device performing the training. Aspects of the generation of the transformed data include the following: Figure 5 and Figure 6 middle.

[0139] At block 750, the parametric model is generated based at least in part on the model training event data. Generating the parametric model can include training a neural network model that approximates a non-parametric transformation from the actual and synthetic event data to the transformed event data. Generating the parametric model can include generating a neural network that approximates the transformation from the generated event data to the transformed generated event data. As discussed above, the construction of the model can be constrained based on factors such as the target particle analyzer or sorting electronics. This can facilitate the generation of a transformation model that can be applied to the target particle analyzer or sorting electronics.

[0140] At block 760, a selection of a target event can be received. For example, the researcher can draw a polygon on the graph to define a range of transformed data values ​​to be sorted. The polygon can define a gate that may be a sorting criterion or associated with a sorting criterion. The selection can be performed using the initial event data transformed by a nonparametric transformation or by a parametric model.

[0141] At block 770, the sorting electronics (e.g., sorting circuitry) may be configured using the parameter model and gating criteria specified at block 760. Configuration of the sorting electronics may include storing the model generated at block 750 in a memory location accessible to the sorting electronics. The model may then be used to transform received event data to evaluate it against the sorting criteria (e.g., the population identified at block 760). As discussed above, the model may be used to configure a field programmable gate array included in the sorting electronics.

[0142] At block 780, the analyzer can use the configured sorting electronics to evaluate and sort the remaining portion of the sample. As new event measurements are collected, the configured sorting electronics can be used to transform the measurements in real time and sort them into designated containers. For example, a deflector plate of the particle analyzer can be activated to direct the desired particles into a designated collection tube.

[0143] Method 700 ends at block 790. However, it will be appreciated that method 700 can be repeated for other events, samples, or experiments. In some embodiments, it may be desirable to retrain the model to accommodate any variation within the sample or to account for changes in the source of the sample. For example, in a therapeutic setting, biological samples may be collected during the administration of a drug or other compound. The model may need to be adjusted to account for the presence of the drug or compound after administration. In such cases, given that the original model has already been trained, the model can be retrained to collect data. This can provide greater resource utilization than training a new model using a corpus of event data collected from the sample.

[0144] The functionality can sort particles based on their membership in a target cluster. Mahalanobis distance can be used to measure the distance between a particle and a known cluster. The sorting decision for a particle can be based on the cluster to which the particle is closest. For certain samples, particles close to a contaminant cluster may be excluded or transferred to an alternate collection container. The criteria for determining closeness can be based on a minimum error probability. The probability can be defined using a configuration or specified as part of the sample experimental setup. Known clusters can be identified using a training set. The training set can include a subset of measurements of particles belonging to a cluster. The mean and covariance of the training set can be generated as a metric representation of a cluster. The metric can then be used as a factor in a Mahalanobis distance equation, which can be evaluated using hardware that can analyze the measurements to generate a timely sorting decision that directs the particles to a specific collection container.

[0145] As used herein, the term "determining" encompasses a variety of activities. For example, "determining" may include calculating, computer computing, processing, deriving, investigating, searching (e.g., searching in a table, database, or another data structure), ascertaining, and the like. Furthermore, "determining" may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory), and the like. Furthermore, "determining" may include resolving, selecting, choosing, establishing, and the like.

[0146] As used herein, the term "providing" encompasses a variety of activities. For example, "providing" may include storing a value at a location on a storage device that facilitates subsequent retrieval, sending a value directly to a recipient via at least one wired or wireless communication medium, sending or storing a reference to a value, etc. "Providing" may also include encoding, decoding, encryption, decryption, validation, verification, etc., performed via hardware components.

[0147] As used herein, the terms "selectively" or "selective" can encompass a variety of activities. For example, a "selective" process can include determining one option from a plurality of options. A "selective" process can include one or more of the following: a dynamically determined input, a pre-configured input, or a user-initiated input for determination. In some embodiments, an n-input switch can be included to provide selective functionality, where n is the number of inputs used to make the selection.

[0148] As used herein, the term "message" encompasses various formats for conveying (e.g., sending or receiving) information. A message can include a machine-readable aggregation of information, such as an XML document, a fixed-field message, a comma-delimited message, or the like. In some embodiments, a message can include a signal for transmitting one or more representations of information. Although described in singular form, it should be understood that a message can be composed, sent, stored, received, or the like in multiple parts.

[0149] As used herein, "user interface" (also referred to as interactive user interface, graphical user interface, or UI) may refer to a web-based user interface that includes data fields, buttons, or other interactive controls for receiving input signals or providing electronic information or providing information to a user in response to any received input signals. TM 、FLASH TM , JAVA TM ,.NET TM 、WINDOWS OS TM , macOS TM , Web services and rich site summary (RSS) and other technologies to implement the UI in whole or in part. In some embodiments, the UI can be included in a standalone client (e.g., a fat client, a thick client) configured to communicate (e.g., send or receive data) according to one or more aspects described.

[0150] As used herein, a "data storage area" may be embodied in a hard drive, solid-state memory, and / or any other type of non-transitory computer-readable storage medium accessible to or by a device, including an access device, a server, or other computing device. As is known in the art, a data storage area may also or alternatively be distributed or partitioned across multiple local and / or remote storage devices without departing from the scope of the present invention. In other embodiments, a data storage area may include or be embodied in a data storage network service.

[0151] As used herein, a phrase referring to "at least one" of a list of items refers to any combination of the items, including individual members. For example, "at least one of the following items: a, b, or c" is intended to include: a, b, c, ab, ac, bc, and abc.

[0152] Those skilled in the art will appreciate that any of a variety of different technologies and techniques may be used to represent information, messages, and signals. For example, references to data, instructions, commands, information, signals, bits, symbols, and chips throughout the specification may be represented by voltages, currents, electromagnetic waves, magnetic fields or particles, light fields or optical particles, or any combination thereof.

[0153] It will be further understood by those skilled in the art that the various illustrative logic blocks, modules, circuits, and algorithmic steps described in conjunction with the embodiments disclosed herein can be used as electronic hardware, computer software, or a combination thereof. In order to clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been generally described above with respect to their functions. Whether such functions are implemented in the form of hardware or software depends on the specific application and the design constraints for the entire system. Technicians can perform the functions in various ways for each specific application, but such execution decisions should not be interpreted as resulting in departure from the scope of the present invention.

[0154] The technology described herein can be implemented in the form of hardware, software, firmware, or any combination thereof. The technology can be implemented in any of a variety of devices, for example, a specially programmed event processing computer, a wireless communication device, or an integrated circuit device. Any function described as a module or component can be executed together in an integrated logic device, or separately in discrete but interoperable logic devices. If implemented in the form of software, the technology can be implemented at least in part by a computer-readable data storage medium, which includes instructions that, when executed, perform one or more of the above methods. The computer-readable data storage medium can constitute part of a computer program product, which can include packaging materials. The computer-readable medium can include a memory or data storage medium, for example, a random access memory (RAM) (e.g., synchronous dynamic random access memory (SDRAM)), a read-only memory (ROM), a non-volatile random access memory (NVRAM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic or optical data storage medium, etc. The computer-readable medium can be a non-temporary storage medium. Alternatively or additionally, the techniques may be implemented at least in part by a computer-readable communication medium that carries or communicates program code in the form of instructions or data structures and that can be accessed, read, and / or executed by a computing device, such as a propagated signal or wave.

[0155] The program code can be executed by a specially programmed transformation processor, which can include one or more processors, such as one or more digital signal processors (DSPs), configurable microprocessors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other equivalent integrated or discrete logic circuit systems. The graphics processor can be specially configured to perform any of the techniques described herein. A combination of computing devices (e.g., a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors combined with a DSP core, or any other such configuration in at least some data connections) can perform one or more of the functions described. In some aspects, the functions described herein can be provided within a dedicated software module or hardware module configured for encoding and decoding, or incorporated into a dedicated transformation control card.

[0156] The methods disclosed herein include one or more steps or actions for implementing the methods. The method steps and / or actions may be interchangeable with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order and / or use of specific steps and / or actions may be modified without departing from the scope of the claims.

[0157] Various embodiments of the invention are described herein. These and other embodiments are within the scope of the following claims.

[0158] Each aspect of the subject matter described above (including embodiments) may be used alone or in combination with one or more aspects or embodiments. Without limiting the content described herein, certain non-limiting aspects of the present invention are provided below. After reading this disclosure, the following will be apparent to those skilled in the art, and each individually numbered content may be used alone or in combination with any preceding or following individually numbered content. This is intended to provide support for all such combinations of the various contents, and is not limited to the combinations of the contents explicitly provided below:

[0159] 1. A method comprising:

[0160] applying a nonparametric transformation to a first raw data set from a flow cytometer to generate a first transformed data set;

[0161] generating a second transformed data set using the first transformed data set; and

[0162] generating a transformation model using the second transformed data set; and

[0163] The transformation model is applied to a second raw data set from the flow cytometer.

[0164] 2. The method of 1, wherein the nonparametric transformation comprises t-distributed stochastic neighbor embedding (tSNE).

[0165] 3. The method of any one of 1-2, wherein the first raw data set comprises data from 10,000 or fewer detected cells of the flow cytometer.

[0166] 4. The method according to any one of 1-3, wherein generating the second transformed data set comprises adding a noise component.

[0167] 5. The method of claim 4, wherein generating the second transformed data set comprises adding the noise component to one or more of the first original data set, the first transformed data set, and a combination thereof.

[0168] 6. The method of claim 5, wherein generating the second transformed data set comprises adding the noise component to a data pair comprising an original data point and a transformed data point.

[0169] 7. The method according to 6, wherein the original data points and the transformed data points are selected by random selection.

[0170] 8. The method according to 7, wherein the original data points and the transformed data points are selected by weighted random selection.

[0171] 9. The method of 8, wherein the random selection is weighted according to one or more of density and cell population.

[0172] 10. The method according to any one of 4-9, wherein the noise component is determined based on a probability distribution.

[0173] 11. The method of 10, wherein the probability distribution is selected from the group consisting of uniform distribution, Gaussian distribution, or Poisson distribution.

[0174] 12. The method according to any one of 1-11, wherein the transformation model is a dynamic algorithm.

[0175] 13. The method of 12, wherein the dynamic algorithm is a machine learning algorithm.

[0176] 14. The method of any one of 1-13, wherein the transformation model is applied to real-time data from the flow cytometer.

[0177] 15. The method of any one of 1-14, further comprising classifying particles in the sample within the fluid medium of the flow cytometer.

[0178] 16. The method of 15 further comprises generating a sorting decision based on the particle classification.

[0179] 17. The method according to any one of 15-16, further comprising sorting one or more particles in the sample.

[0180] 18. The method according to any one of 1-17, wherein the method is performed by an integrated circuit device.

[0181] 19. The method of 18, wherein the integrated circuit device is a field programmable gate array (FPGA).

[0182] 20. The method of 18, wherein the integrated circuit device is an application specific integrated circuit (ASIC).

[0183] 21. The method of 18, wherein the integrated circuit device is a complex programmable logic device (CPLD).

[0184] 22. A computer-implemented method comprising:

[0185] controlled by one or more processing devices,

[0186] receiving, from a particle analyzer, measurements of a first portion of particles associated with the experiment;

[0187] converting the measurements into initial transformed measurements using a nonparametric transformation;

[0188] generating a parametric model to receive as input the measurements of the first portion of particles associated with the experiment and to generate as output the initial transformed measurements; and

[0189] The particle analyzer is configured to:

[0190] converting measurements of particles included in a second portion of the particles based at least in part on the parametric model; and

[0191] The particles are sorted based at least in part on the measurements and the sorting criteria.

[0192] 23. The computer-implemented method of 22, further comprising:

[0193] Gate information is received, the gate information identifying a range of measurement values ​​in a non-parametric space for classifying the particles, wherein the sorting criteria comprises the gate information.

[0194] 24. The computer-implemented method of any one of 22-23, further comprising:

[0195] transmitting the parametric model to the particle analyzer; and

[0196] The particle analyzer is caused to configure sorting circuitry based at least in part on the parametric model and the sorting criteria.

[0197] 25. The computer-implemented method of 24, wherein the sorting circuit comprises a field programmable gate array.

[0198] 26. The computer-implemented method of any one of 22-25, wherein the measurements received from the particle analyzer include measurements of light emitted by the first portion of particles in the form of fluorescence.

[0199] 27. The computer-implemented method of 26, wherein the light emitted in the form of fluorescence by the first portion of particles comprises light emitted in the form of fluorescence by antibodies bound to the first portion of particles.

[0200] 28. The computer-implemented method of any one of 22-27, further comprising:

[0201] Select the measured values ​​of the particle and the initial transformed measured values; and

[0202] Generating a composite measurement pair based at least in part on:

[0203] (i) a measurement value of the particle,

[0204] (ii) an initial transformed measurement of the particle; and

[0205] (iii) noise value, wherein the integrated measurement value pair includes an integrated

[0206] measurements and integrated transformed measurements; and

[0207] The parameter model further receives the integrated measurement value as input and generates the integrated transformed measurement value as output.

[0208] 29. The computer-implemented method of 28, further comprising:

[0209] receiving gate information identifying a range of measurement values ​​for classifying the particle, and

[0210] Wherein the particles are selected based at least in part on the measurement value corresponding to the range of measurement values.

[0211] 30. The computer-implemented method of any one of 22-29, further comprising:

[0212] receiving a hardware identifier from the particle analyzer, the hardware identifier indicating sorting circuitry implemented in the particle analyzer;

[0213] determining an architecture of the parametric model based at least in part on the hardware identifier, wherein the architecture of the parametric model is applicable to the sorting circuit; and

[0214] Generating the parameter model includes generating the parameter model consistent with the architecture.

[0215] 31. The computer-implemented method of 30, wherein the parametric model comprises a neural network, and wherein the architecture identifies the number of layers of the neural network and the number of nodes in each layer.

[0216] 32. The computer-implemented method of any one of 22-31, wherein the parametric model comprises a neural network, and wherein generating the parametric model comprises:

[0217] assigning initial weights to connections between a first node in a first layer of the parametric model and a second node in a second layer of the parametric model;

[0218] generating output suitable for particle measurements;

[0219] comparing the output to the transformed measurements of the particles; and

[0220] adjusting the initial weights based at least in part on a difference between the output and the transformed measurement,

[0221] The subsequent difference value of the particle is smaller than the difference value after adjusting the initial weight.

[0222] 33. A system comprising a processor, the processor comprising a memory operatively coupled thereto, wherein the memory comprises instructions stored thereon that, when executed by the processor, cause the processor to:

[0223] applying a nonparametric transformation to a first raw data set from a flow cytometer to generate a first transformed data set;

[0224] generating a second transformed data set using the first transformed data set; and

[0225] generating a transformation model using the second transformed data set; and

[0226] The transformation model is applied to a second raw data set from the flow cytometer.

[0227] 34. A system according to 33, wherein the non-parametric transformation comprises t-distributed stochastic neighbor embedding (tSNE).

[0228] 35. The system of any one of claims 33-34, wherein the first raw data set comprises data from 10,000 or fewer detected cells of the flow cytometer.

[0229] 36. A system according to any one of 33-35, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component.

[0230] 37. A system according to 36, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to one or more of the first original data set, the first transformed data set, and a combination thereof.

[0231] 38. A system according to 37, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to data pairs (comprising original data points and transformed data points).

[0232] 39. The system of claim 38, wherein the original data points and the transformed data points are selected by random selection.

[0233] 40. The system of claim 39, wherein the original data points and the transformed data points are selected by weighted random selection.

[0234] 41. A system according to 40, wherein the random selection is weighted by one or more of density and cell population.

[0235] 42. A system according to any one of 36-41, wherein the noise component is determined based on a probability distribution.

[0236] 43. A system according to 42, wherein the probability distribution is selected from the group consisting of a uniform distribution, a Gaussian distribution, or a Poisson distribution.

[0237] 44. A system according to any one of 33-43, wherein the transformation model is a dynamic algorithm.

[0238] 45. A system according to 44, wherein the dynamic algorithm is a machine learning algorithm.

[0239] 46. ​​A system according to any one of 33-45, wherein the transformation model is applied to real-time data from the flow cytometer.

[0240] 47. A system according to any one of 33-46, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to classify particles in a sample within the fluid medium of the flow cytometer.

[0241] 48. A system according to claim 47, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate a sorting decision based on the particle classification.

[0242] 49. The system of any one of 33-48, further comprising an integrated circuit device.

[0243] 50. The system of claim 49, wherein the integrated circuit device is a field programmable gate array (FPGA).

[0244] 51. The system of claim 49, wherein the integrated circuit device is an application specific integrated circuit (ASIC).

[0245] 52. The system of claim 49, wherein the integrated circuit device is a complex programmable logic device (CPLD).

[0246] 53. A system comprising:

[0247] one or more processing devices; and

[0248] A computer-readable storage medium containing instructions that, when executed by the one or more processing devices, cause the system to:

[0249] receiving, from a particle analyzer, measurements of a first portion of particles associated with the experiment;

[0250] converting the measurements into initial transformed measurements using a nonparametric transformation;

[0251] generating a parametric model to receive as input raw measurements of the first portion of particles associated with the experiment and to generate as output the initial transformed measurements;

[0252] The particle analyzer is configured to:

[0253] converting measurements of particles included in the second portion of the particles based at least in part on the parametric model;

[0254] generating a control signal to adjust an operating state of an analyzing device included in the particle analyzer; and

[0255] The control signal is transmitted to the analysis device to achieve the operating state.

[0256] 54. The system of claim 53, wherein the analysis device comprises sorting electronics communicatively coupled to the deflection plates, and

[0257] wherein the particle analyzer is further configured to identify a target container suitable for particles based at least in part on the converted measurement value corresponding to a sorting criterion associated with the target container, and

[0258] Adjusting the operating state includes applying an electric charge through the deflection plate to guide the particles into a target container.

[0259] 55. The system of 53, wherein the analytical device comprises a fluidics system, and

[0260] Adjusting the operating state includes adjusting the pressure applied during the experiment.

[0261] 56. A system comprising:

[0262] a light source configured to irradiate a sample (including particles) in a fluid medium;

[0263] a light detection system comprising a photodetector configured to generate a data signal using light detected from particles within the fluid medium; and

[0264] A processor comprising a memory operatively coupled thereto, wherein the memory comprises instructions stored thereon that, when executed by the processor, cause the processor to:

[0265] applying a nonparametric transform to a first set of data signals from the light detection system to generate a first transformed data set;

[0266] generating a second transformed data set using the first transformed data set; and

[0267] generating a transformation model using the second transformed data set; and

[0268] The transformation model is applied to a second set of data signals from the light detection system.

[0269] 57. A system according to 56, wherein the non-parametric transformation comprises t-distributed stochastic neighbor embedding (tSNE).

[0270] 58. The system of any one of claims 56-57, wherein the first set of data signals from the light detection system comprises data from 10,000 or fewer detected cells from the flow cytometer.

[0271] 59. A system according to any one of 56-58, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component.

[0272] 60. A system according to 59, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to one or more of the first set of data signals from the optical detection system, the first transformed data set, and a combination thereof.

[0273] 61. A system according to 60, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to data pairs (comprising original data points and transformed data points).

[0274] 62. A system according to 61, wherein the original data points and the transformed data points are selected by random selection.

[0275] 63. A system according to 62, wherein the original data points and the transformed data points are selected by weighted random selection.

[0276] 64. A system according to 63, wherein the random selection is weighted by one or more of density and cell population.

[0277] 65. A system according to any one of 59-64, wherein the noise component is determined based on a probability distribution.

[0278] 66. A system according to 65, wherein the probability distribution is selected from the group consisting of uniform distribution, Gaussian distribution or Poisson distribution.

[0279] 67. A system according to any one of 56-66, wherein the transformation model is a dynamic algorithm.

[0280] 68. A system according to 67, wherein the dynamic algorithm is a machine learning algorithm.

[0281] 69. A system according to any of 56-68, wherein the transformation model is applied to a real-time data signal from the light detection system.

[0282] 70. A system according to any one of 56-69, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to classify particles in the sample within the fluid medium.

[0283] 71. A system according to 70, wherein the processor includes a memory operably coupled to itself, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate a sorting decision based on the particle classification.

[0284] 72. The system according to any one of 56-71 further comprises a cell sorter.

[0285] 73. A system according to 72, wherein the cell sorter comprises a droplet deflector.

[0286] 74. The system of any one of 56-73, further comprising an integrated circuit device.

[0287] 75. The system of 74, wherein the integrated circuit device is a field programmable gate array (FPGA).

[0288] 76. The system of 74, wherein the integrated circuit device is an application specific integrated circuit (ASIC).

[0289] 77. The system of 74, wherein the integrated circuit device is a complex programmable logic device (CPLD).

[0290] 78. An integrated circuit programmed to:

[0291] applying a nonparametric transformation to a first raw data set from a flow cytometer to generate a first transformed data set;

[0292] generating a second transformed data set using the first transformed data set; and

[0293] generating a transformation model using the second transformed data set; and

[0294] The transformation model is applied to a second raw data set from the flow cytometer.

[0295] 79. An integrated circuit according to claim 78, wherein the non-parametric transformation comprises t-distributed stochastic neighbor embedding (tSNE).

[0296] 80. An integrated circuit according to any one of 78-79, wherein the first raw data set comprises data from 10,000 or fewer detected cells of the flow cytometer.

[0297] 81. An integrated circuit according to any one of 78-80, wherein the integrated circuit is programmed to generate the second transformed data set by adding a noise component.

[0298] 82. The integrated circuit of claim 78, wherein the integrated circuit is programmed to generate the second transformed data set by adding a noise component to one or more of the first original data set, the first transformed data set, and a combination thereof.

[0299] 83. An integrated circuit according to claim 82, wherein the integrated circuit is programmed to generate the second transformed data set by adding a noise component to data pairs (comprising original data points and transformed data points).

[0300] 84. An integrated circuit according to 83, wherein the original data points and the transformed data points are selected by random selection.

[0301] 85. An integrated circuit according to 84, wherein the original data points and the transformed data points are selected by weighted random selection.

[0302] 86. An integrated circuit according to 85, wherein the random selection is weighted by one or more of density and cell population.

[0303] 87. An integrated circuit according to any one of 81-86, wherein the noise component is determined based on a probability distribution.

[0304] 88. The integrated circuit of 87, wherein the probability distribution is selected from the group consisting of a uniform distribution, a Gaussian distribution, or a Poisson distribution.

[0305] 89. An integrated circuit according to any one of 78-88, wherein the transformation model is a dynamic algorithm.

[0306] 90. The integrated circuit of 89, wherein the dynamic algorithm is a machine learning algorithm.

[0307] 91. An integrated circuit according to any of 78-90, wherein the transformation model is applied to real-time data from the flow cytometer.

[0308] 92. An integrated circuit according to any one of 78-91, wherein the integrated circuit is programmed to classify particles in a sample within a fluid medium of the flow cytometer.

[0309] 93. An integrated circuit according to 92, wherein the integrated circuit is programmed to generate a sorting decision based on the particle classification.

[0310] 94. An integrated circuit according to any one of 78-93, wherein the integrated circuit device is a field programmable gate array (FPGA).

[0311] 96. An integrated circuit according to any of 78-93, wherein the integrated circuit device is an application specific integrated circuit (ASIC).

[0312] 97. An integrated circuit according to any one of 78-93, wherein the integrated circuit device is a complex programmable logic device (CPLD).

[0313] Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be apparent to one skilled in the art in view of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims.

[0314] Therefore, the foregoing merely illustrates the principles of the present invention. It should be understood that those skilled in the art will be able to design various structures, which, although not explicitly stated or shown here, reflect the principles of the present invention and do not exceed the spirit and scope of the present invention. In addition, all examples and conditional language listed herein are primarily intended to help the reader understand the principles of the present invention and the ideas provided by the inventors to further expand this field, and should be interpreted as not being limited by these specifically listed examples and conditions. Moreover, all statements herein citing the principles, aspects and embodiments of the present invention and specific examples thereof are intended to cover their equivalents in structure and function. In addition, the equivalents are intended to include currently known equivalents and equivalents to be developed in the future, that is, any elements developed that have the same function and do not consider their structure. Moreover, regardless of whether the disclosure of the present invention is explicitly stated in the claims, none of the contents thereof will be disclosed to the public.

[0315] Therefore, the scope of the present invention is not limited to the exemplary embodiments shown and described herein. Rather, the scope and spirit of the present invention are embodied by the appended claims. In the claims, 35 U.S.C. § 112(f) or 35 U.S.C. § 112(6) is expressly defined as being cited only when the phrase "means for..." or "step for..." is explicitly used at the beginning of the claim limitation; 35 U.S.C. § 112(f) or 35 U.S.C. § 112(6) is not cited if the phrase is not used in the claim limitation.

Claims

1. A system comprising: one or more processing devices; and A computer-readable storage medium containing instructions that, when executed by the one or more processing devices, cause the system to: receiving, from a particle analyzer, measurements of a first portion of particles associated with the experiment; converting the measurements to initial transformed measurements using a nonparametric transformation; generating a parametric model to receive as input raw measurements of the first portion of particles associated with the experiment and to generate as output the initial transformed measurements; The particle analyzer is configured as follows: converting measurements of particles included in the second portion of particles based at least in part on the parametric model; generating a control signal to adjust an operating state of an analysis device included in the particle analyzer; as well as The control signal is transmitted to the analysis device to achieve the operating state.

2. The system of claim 1, wherein the analysis device comprises sorting electronics communicatively coupled to the deflection plates, and wherein the particle analyzer is further configured to identify a target container for the particles based at least in part on the converted measurements corresponding to a sorting criterion associated with the target container, and Adjusting the operating state includes applying an electric charge through the deflection plate to guide the particles into a target container.

3. The system of claim 3, wherein the analysis device comprises a fluidics system, and Adjusting the operating conditions includes adjusting the pressure applied during the experiment.

4. A system comprising: a light source configured to irradiate a sample containing particles in a fluid medium; a light detection system including a photodetector configured to generate a data signal using light detected from particles within the fluid medium; as well as A processor comprising a memory operably coupled to the processor, wherein the memory comprises instructions stored thereon, which when executed by the processor cause the processor to perform the following operations: applying a non-parametric transform to a first set of data signals from the light detection system to generate a first transformed data set; generating a second transformed data set using the first transformed data set; as well as generating a transformation model using the second transformed data set; as well as The transformation model is applied to a second set of data signals from the light detection system.

5. The system of claim 5, wherein the non-parametric transformation comprises t-distributed stochastic neighbor embedding (tSNE).

6. The system of any one of claims 4-5, wherein the first set of data signals from the light detection system comprises data from 10,000 or fewer detected cells from a flow cytometer.

7. A system according to any one of claims 4-6, wherein the processor includes a memory operably coupled to the processor, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component.

8. The system of claim 7 , wherein the processor comprises a memory operably coupled to the processor, wherein the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to one or more of the first set of data signals from the optical detection system, the first transformed data set, and a combination thereof.

9. A system according to claim 8, wherein the processor includes a memory operably coupled to the processor, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to generate the second transformed data set by adding a noise component to a data pair comprising an original data point and a transformed data point.

10. The system of claim 9, wherein the original data points and the transformed data points are selected by random selection.

11. The system of claim 10, wherein the original data points and the transformed data points are selected by weighted random selection.

12. The system of claim 11, wherein the random selection is weighted by one or more of density and cell population.

13. The system according to any one of claims 9-12, wherein the noise component is determined based on a probability distribution.

14. The system of claim 13, wherein the probability distribution is selected from the group consisting of a uniform distribution, a Gaussian distribution, or a Poisson distribution.

15. The system according to any one of claims 4-14, wherein the transformation model is a dynamic algorithm.

16. The system of claim 15, wherein the dynamic algorithm is a machine learning algorithm.

17. The system of any one of claims 4-16, wherein the transformation model is applied to real-time data signals from the light detection system.

18. A system according to any one of claims 4-17, wherein the processor includes a memory operably coupled to the processor, wherein the memory includes instructions stored thereon, which, when executed by the processor, cause the processor to classify particles in the sample within the fluid medium.

19. The system of claim 18, wherein the processor comprises a memory operably coupled to the processor, wherein the memory includes instructions stored thereon that, when executed by the processor, cause the processor to generate a sorting decision based on the particle classification.

20. The system according to any one of claims 4-19, further comprising a cell sorter.

21. The system of claim 20, wherein the cell sorter comprises a droplet deflector.

22. The system of any one of claims 4-21, further comprising an integrated circuit device.

23. The system of claim 22, wherein the integrated circuit device is a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or a complex programmable logic device (CPLD).

Images