Systems and methods for quantifying the presence of a target organism in a sample

Abstract

In some examples, a system for amplifying and quantifying a target organism present in a sample includes a detection device configured to amplify and detect nucleic acids associated with the target organism. The detection device is configured to receive a sample and amplify nucleic acids in the sample within an amplification cycle. The detection device is configured to capture a data set including measurements of nucleic acids collected during the amplification cycle. The system also includes a computing device configured to receive the data set and apply a machine learning system to the data set. The machine learning system is trained to estimate the amount of the target organism present in the sample based on the measurements in the data set.

Description

Technical Field

[0001] The present disclosure relates to systems and methods for detecting a target organism, and in particular, to systems and methods for estimating the amount of a target organism. Background Art

[0002] Foodborne bacterial infections and diseases are a persistent threat to public health. Regulatory agencies such as the U.S. Department of Agriculture's Food Safety and Inspection Service respond to this threat by issuing pathogen reduction performance standards for pathogens (e.g., Salmonella and Campylobacter) in food, feed, water, and corresponding processing environments. Some such pathogen reduction standards apply presence / absence criteria, while other standards require quantitative information about pathogens.

[0003] Food, feed and water producers use quantitative techniques to measure the amount of microorganisms (such as bacterial pathogens) in food, feed (such as animal feed), water and corresponding processing environments. Such producers can, for example, quantify total bacteria and indicator bacteria to assess the effectiveness of pathogen intervention processes (such as food safety procedures based on hazard analysis and critical control points (HACCP) and other sanitation control measures). Generally, people seeking to measure the amount of pathogens rely on traditional methods for quantification, such as the maximum possible number (MPN) estimate based on continuous culture dilution. Such methods are usually time-consuming, cumbersome and prone to error. In addition, such methods may require special culture media, and may require 24 hours or longer to give results. Nevertheless, food, feed and water producers still rely on these methods to quantify total bacteria and indicator organisms (such as Escherichia coli (E.coli or coliform)). Summary of the invention

[0004] The present disclosure provides systems and methods for quantifying one or more target organisms (such as one or more species of a bacterial genus) present in a biological assay (e.g., a specific sample of food, feed, water, raw materials, or corresponding environmental samples) using nucleic acid amplification assays, and systems and methods for training a machine learning system to quantify the target organisms present in the biological assay. The present disclosure also provides methods for training a machine learning system to quantify the target organisms present in an inhibited biological assay.

[0005] The exemplary system includes a detection device configured to amplify and detect a target nucleic acid associated with a target organism, such as a thermal cycler configured to perform qPCR or other types of PCR. Some other such detection devices may be isothermal devices configured to perform loop-mediated isothermal DNA amplification (LAMP). The detection device includes a reaction chamber configured to receive a sample having an amount of target nucleic acid and amplify the target nucleic acid in the sample during a nucleic acid amplification cycle; and a detector configured to capture a measurement result representing the amount of target nucleic acid present in the sample during the nucleic acid amplification cycle, and store the measurement result in a data set, wherein the data set includes: a first data subset, the first data subset including the time T max The measurement results obtained before, where time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; a second data subset, the second data subset comprising the measurement results taken after the first time point in the nucleic acid amplification cycle but before a second time point, the second time point occurring at T max Thereafter; and a third data subset comprising measurements taken after a second time point in the nucleic acid amplification cycle.

[0006] The system also includes a machine learning system configured to receive the first data subset, the second data subset, and the third data subset, and apply the machine learning system to the data subsets. In some examples, the first data subset, the second data subset, and the third data subset include all measurements in the data set. The machine learning system is trained to estimate the amount of the target organism present in the sample based on the measured samples in the first data subset, the second data subset, and the third data subset.

[0007] An exemplary method includes receiving a plurality of data sets, wherein each data set is associated with a biological assay, each data set comprising measurements of a target nucleic acid detected within the associated biological assay, the measurements being performed on the associated biological assay by a nucleic acid amplification device of a specified type and collected within at least a portion of a nucleic acid amplification cycle, wherein the target nucleic acid is associated with a target organism; labeling each data set with an estimate of the amount of the target organism present within the associated biological assay; and training a machine learning system with the labeled data sets to estimate the amount of the target organism within the biological assay based on testing performed on the target nucleic acid in the biological assay by the nucleic acid amplification device of the specified type.

[0008] An exemplary non-transitory computer-readable medium includes instructions that, when executed by a processing circuit, cause the processing circuit to: receive a data set generated by amplifying an amount of nucleic acid in a sample during a nucleic acid amplification cycle, wherein the nucleic acid is associated with a target organism, the data set including measurements collected during the nucleic acid amplification cycle representing the amount of nucleic acid in the sample, wherein the data set includes: a first data subset including at time T max The measurement results obtained before, where time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; a second data subset, the second data subset comprising the measurement results taken after the first time point in the nucleic acid amplification cycle but before a second time point, the second time point occurring at T max thereafter; and a third data subset comprising measurements taken after a second time point in the nucleic acid amplification cycle; and applying a machine learning system to the data subset, wherein the machine learning system is trained to estimate an amount of a target organism present in the sample based on the measurements present in the first data subset, the second data subset, and the third data subset.

[0009] An exemplary method for training a machine learning system to quantify a target organism present in a biological assay includes receiving data sets, each data set associated with a biological assay, each data set comprising data collected by a detector during nucleic acid amplification of a target nucleic acid in the associated biological assay within one or more nucleic acid amplification cycles, wherein the data collected by the detector comprises activity measurements taken at different times during the one or more nucleic acid amplification cycles, wherein the target nucleic acid is associated with a target organism, and wherein the biological assays include biological assays with different inhibition levels; labeling each data set with an estimate of the amount of the target organism present in the associated biological assay; and training the machine learning system to estimate the amount of the target organism in a selected biological assay, the training being based on the activity measurements stored in each data set and the estimate of the amount of the target organism present in the biological assay associated with each respective data set.

[0010] An exemplary system for quantifying a target organism present in a sample includes a detection device configured to amplify and detect a target nucleic acid associated with the target organism, the detection device including a reaction chamber configured to receive a bioassay having an amount of the target nucleic acid and amplify the target nucleic acid in the sample during a nucleic acid amplification cycle, and a detector configured to capture activity measurements during the nucleic acid amplification cycle that represent the amount of the target nucleic acid present in the sample taken at different times during the nucleic acid amplification cycle. The system also includes a machine learning system configured to receive the measurements and apply the machine learning system to the measurements, wherein the machine learning system is trained using bioassays having different inhibition levels to estimate the amount of the target organism present in the sample based on the measurements, wherein the training includes training the machine learning system to estimate the amount of the target organism within the selected bioassays, the training being based on the activity measurements stored in each data set and the estimate of the amount of the target organism present in the bioassay associated with each respective data set.

[0011] Therefore, in the systems and methods described herein, machine learning systems (such as support vector machines, enhanced decision trees, neural networks and / or other systems) can be used to collect and analyze data generated by biological assays. Such data can be used to train and build machine learning systems for specific pathogens. A machine learning system trained with one or more appropriate data sets can check most or all of the signal response in a molecular diagnostic assay (e.g., qPCR and / or LAMP). Therefore, such machine learning systems can be used to extract nonlinear relationships between variables and to estimate the amount of organisms present in the original sample. The quantification of pathogens can be achieved by applying a trained machine learning system to such molecular methods, and results can be produced in a shorter time period than traditional methods, and / or more accurate results can be provided at a lower cost relative to molecular methods that do not include the application of such trained machine learning systems.

[0012] The details of one or more examples will be set forth in the accompanying drawings and the description below.Other features, objects, and advantages of the disclosure will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 is a block diagram illustrating an exemplary system including a nucleic acid amplification device configured to amplify and detect nucleic acid associated with a target organism and a user device configured to estimate the amount of the target organism according to one aspect of the present disclosure.

[0014] Figure 2 is a block diagram showing an exemplary system according to one aspect of the present disclosure, the system including an external device such as a server, and a Figure 1 The access point of the nucleic acid amplification device.

[0015] Figure 3 The present invention is a diagram showing an aspect of the present invention. Figure 1 Schematic conceptual diagram of an exemplary user equipment.

[0016] Figure 4 is a flow chart illustrating exemplary points for pathogen testing before, during, and / or after food or feed production according to one aspect of the present disclosure.

[0017] Figure 5 is a flow chart illustrating an exemplary technique for estimating the amount of a target organism in a sample according to one aspect of the present disclosure.

[0018] Figure 6 Real-time detection of nucleic acid amplification during LAMP amplification cycles based on measurements of bioluminescence intensity over time according to one aspect of the present disclosure is shown.

[0019] Figure 7 is a schematic diagram illustrating representative features of an exemplary qPCR technique according to one aspect of the present disclosure.

[0020] Figure 8 The limitations of the standard curve method for quantifying pathogens when using cell counting according to one aspect of the present disclosure are shown.

[0021] 9A to 9C is a flow chart illustrating an exemplary technique for training a machine learning system and for using the trained machine learning system to estimate an initial amount of a target organism in a sample according to one aspect of the present disclosure.

[0022] Fig.10 is a block diagram illustrating a device training system according to one aspect of the present disclosure.

[0023] Fig.11 Techniques for training a machine learning model to estimate cell counts of target cells seeded into a matrix, and techniques for using the trained machine learning model to estimate cell counts in a matrix based on the trained model, are shown according to one aspect of the present disclosure.

[0024] Fig.12 Shown are the logarithmic differences between cell count estimates made by a trained machine learning system and different cell counts of Salmonella cells inoculated into a poultry rinse matrix according to one aspect of the present disclosure.

[0025] Fig.13Shown are the logarithmic differences between cell count predictions made by a trained machine learning system and different cell counts of Salmonella cells inoculated into poultry rinse matrix and a 1:10 dilution of poultry rinse matrix according to one aspect of the present disclosure.

[0026] Fig.14 Various metrics for measuring the performance of regression for cell count prediction using various machine learning techniques in accordance with aspects of the present disclosure are shown.

[0027] Fig.15 is a conceptual diagram illustrating nucleic acid amplification in a standard sample and an inhibited sample during a LAMP amplification cycle according to one aspect of the present disclosure.

[0028] Fig.16 is a flow chart illustrating an exemplary technique for training a machine learning system to quantify a target organism in an inhibited sample according to one aspect of the present disclosure. DETAILED DESCRIPTION

[0029] In the following discussion, the term "food" also includes beverages. The term "water" includes drinking water, but the term "water" also includes water used in other situations where quantitative measurement results of one or more of the microorganisms in the water are required.

[0030] As mentioned above, food, feed and water producers use quantitative techniques to measure the amount of microorganisms (such as bacterial pathogens) in food, feed (e.g., animal feed), water and corresponding processing environments. Quantitative techniques are used, for example, to assess the effectiveness of pathogen intervention processes used during food production. Such analysis can lead to more effective risk analysis and lead to the development of more effective ways to reduce the level of pathogens in food, feed and water supplies. However, the traditional methods discussed above for measuring the amount of pathogens in bioassays are time-consuming, cumbersome and prone to error. They may require specialized culture media and may take one or more days to give results.

[0031] Molecular methods (such as LAMP or PCR) can also be used to quantify pathogens extracted from samples. Molecular methods for pathogen quantification provide results in a shorter amount of time (such as in hours, rather than one or more days) than more traditional methods. In addition, they are not limited to quantifying total bacteria and indicator bacteria, but can also be used to quantify specific bacteria, yeast, mold or other pathogens. In implementation, the producer determines the amount by extrapolating the amount of pathogens in the sample from a standard curve constructed by known nucleic acid concentrations based on the test results from the sample. However, the standard curve constructed by known nucleic acid concentrations may not correspond well to the count of organisms in samples collected from, for example, a production environment.

[0032] For example, qPCR is widely used as a molecular method for detecting a variety of bacteria. qPCR can also be used to absolutely quantify the pathogens present in a given amount of sample. A standard curve containing a known amount of target DNA (plasmid, genomic DNA or other nucleic acid molecules) is run in parallel with an unknown sample. Based on the standard curve, the reaction efficiency and dilution steps for nucleic acid extraction and analysis, the absolute number of pathogens in an unknown sample can be estimated. In these types of analyses, a linear regression model is used, amplification efficiency becomes critical, and each run requires running standards, thereby increasing cost, time, and possible sample contamination. In addition, when cell counting (rather than DNA) is used, the use of the standard curve method is limited. For these reasons, traditional methods are more preferred than molecular methods for microbial quantification.

[0033] As mentioned above, the determination based on the molecular method such as nucleic acid amplification (for example, LAMP or PCR) is efficient. However, they can be affected by the presence of matrix-derived substances, which can interfere with or prevent the reaction from being correctly carried out, and this process is called inhibition. In food production, matrix-derived substances (such as spices and environmental samples) can serve as inhibitors that can interfere with nucleotide amplification determinations (such as PCR and LAMP), thereby causing false negative results.

[0034] It may be difficult to eliminate inhibition. Careful sample handling can be used, for example, to remove inhibitory substances. However, sample handling cannot be relied upon to completely remove inhibitory substances.

[0035] Amplification controls can also be used to control inhibition. Such controls can be used, for example, to verify whether the assay has been correctly performed. Typically, an internal amplification control (IAC) is a non-target DNA sequence present in a reaction identical to a sample or target nucleic acid extract. If it is successfully amplified to produce a signal, it is considered that any non-production of a target signal in the reaction indicates that the sample does not contain a target pathogen or organism. However, if the reaction neither produces a signal from the target nor a signal from the IAC, it indicates that the reaction has failed, indicating that the target organism does not exist when the target organism is in fact present (i.e., "false negative"). Therefore, detecting false negatives during the amplification cycles may be critical for reliable testing.

[0036] The interpolation of amplification controls increases the complexity and cost of molecular methods. When applying molecular methods to detect or quantify target organisms in samples, even in the face of inhibition, the use of amplification controls will be advantageous. Therefore, methods for identifying and correcting inhibition are provided below. These methods can, for example, be used to correct quantification in nucleotide amplification without the need for internal or external amplification controls.

[0037] The following disclosure describes systems and methods for quantifying pathogens in biological assays. The following disclosure also describes systems and methods for training and using machine learning systems in molecular methods for pathogen quantification, thereby improving the accuracy of pathogen quantification and reducing or eliminating the need to prepare and use standard curves for each run. In some exemplary methods described herein, LAMP bioluminescence assays and / or PCR assays (e.g., qPCR assays) can be used in training runs to amplify target nucleic acids (e.g., nucleic acids associated with target organisms) present in a sample with a known initial amount and detect light generated in the sample during target nucleic acid amplification. In other exemplary methods described herein, assays such as nickase amplification reactions (NEAR), helicase-dependent amplification (HDA), nucleic acid sequence-based amplification (NASBA), or transcription-mediated amplification (TMA) assays can be used.

[0038] Any suitable variant of such determination can be used.The variant of the operable traditional LAMP determination can include colorimetric LAMP (cLAMP) determination, wherein the pH variation driven by proton accumulation during LAMP can be visualized via observing the color change of the pH-sensitive colorimetric dye that occurs with nucleic acid amplification.Other such variants can include turbidity-LAMP determination, wherein the formation of magnesium pyrophosphate during LAMP causes turbidity, and this turbidity increases in association with nucleic acid yield and can be quantitative in real time.The materials and methods used in such variants of traditional LAMP determination and / or PCR determination can be understood by those skilled in the art, therefore no longer described in detail herein.It should be understood that exemplary nucleic acid amplification techniques and variants thereof described herein are not intended to be limited.On the contrary, any suitable nucleic acid amplification techniques can be used in technology described herein, such as in a training run for amplifying target nucleic acids.

[0039] Data from the training run can be fed into the machine learning system to train the machine learning system. The trained machine learning system can then be used to estimate the unknown initial amount of the target organism present in a sample (such as a food sample, a feed sample, water, or an environmental sample from a food or feed processing environment). In other exemplary methods described herein, LAMP bioluminescence assays and / or PCR assays (e.g., qPCR assays) can be used to amplify the target nucleic acid (e.g., nucleic acid associated with the target organism) in a series of samples with a known initial amount of the target organism. The method collects data representing the light generated in the sample during the target nucleic acid amplification for each sample, and the collected data is associated with a known amount of target nucleic acid or with a known amount of detected organisms. The data from the training run is then fed into the machine learning system to train the machine learning system. The trained machine learning system can then be used to estimate the unknown initial amount of the target organism present in a sample (such as a food sample, a feed sample, water, or an environmental sample from a food or feed processing environment).

[0040] In other exemplary methods described herein, LAMP bioluminescence assays and / or PCR assays (e.g., qPCR assays) can be used to obtain data corresponding to samples collected from a specific environment (e.g., a poultry processing plant or a cheese factory). The samples are viewed using traditional quantitative methods, and each sample is labeled with a quantity value determined by one or more of the traditional methods. The data from the labeled samples are then fed into a machine learning system to train the machine learning system for the specific environment. The trained machine learning system can then be used to better estimate the unknown initial amount of target organisms and / or nucleic acids present in a sample (such as a food sample, a feed sample, water, or an environmental sample from a specific environment).

[0041] It should be noted that while in some examples, the nucleic acid associated with the target organism may be described herein as DNA, in other examples, the nucleic acid associated with the target organism may be RNA. In such other examples, amplification techniques on total RNA or mRNA of a sample, such as quantitative reverse transcription PCR (RT-qPCR) and reverse transcription LAMP (RT-LAMP), can be used to train a machine learning system to estimate the initial amount of a target organism in a sample and / or to apply such a trained machine learning system.

[0042] Each machine learning system is based on at least one model. The model can be a regression model based on techniques such as support vector regression, random forest regression, linear regression, ridge regression, logistic regression, Lasso regression, or nearest neighbor regression. Or the model can be a classification model based on techniques such as support vector machines, decision trees and random forests, linear discriminant analysis, neural networks, nearest neighbor classifiers, stochastic gradient descent classifiers, Gaussian process classification, or original Bayes. Both types of models rely on the use of labeled data sets to train the model.

[0043] Figure 1 1 is a block diagram illustrating an exemplary system including a nucleic acid amplification device configured to amplify and detect nucleic acids associated with a target organism and a user device configured to estimate the amount of the target organism according to one aspect of the present disclosure. According to one aspect of the present disclosure, the nucleic acid amplification device 8 is configured to amplify and detect the target nucleic acid. The nucleic acid amplification device 8 includes a reaction chamber 10 configured to amplify the target nucleic acid. In an exemplary method, as Figure 1 As shown, the reaction chamber 10 includes a block 12 that can be heated and / or cooled via a heat source (such as a Peltier system). Figure 1 As shown, the block 12 defines a plurality of recesses 14, each of which may be sized to receive a reaction vessel, which may be any suitable plastic tube configured for nucleic acid amplification assays. The nucleic acid amplification apparatus 8 also includes a detector 16 and a control unit 18. The detector 16 may be configured to capture light within the reaction chamber 10 under the control of the control unit 18. For example, the detector 16 may be configured to capture a data set comprising a time series measurement sample of light emitted by a luminescent substance within a sample contained within a reaction vessel received within one of the recesses 14 during one or more nucleic acid amplification cycles. In some examples, the sample may include a target nucleic acid and a luminescent substance, the latter of which may emit light in a stoichiometric relationship with the target nucleic acid, such that the light emitted by the luminescent substance increases as the amount of replicated target nucleic acid in the sample increases.

[0044] In some examples, the nucleic acid amplification device 8 can be any suitable nucleic acid amplification device configured for LAMP (e.g., a traditional LAMP assay or cLAMP, turbidity LAMP, or other variations of a traditional LAMP assay). In examples where light is emitted by a luminescent substance captured by the detector 16, the light can be bioluminescence, fluorescence, or light of any visible color. In examples using turbidity LAMP technology, the detector can measure at least one of absorbance, transmittance, or reflectance. In addition or alternatively, the nucleic acid amplification device 8 can be any suitable nucleic acid amplification device configured for qPCR or any other nucleic acid amplification technology (e.g., NEAR, HDA, NASBA, TMA, or others). In some such other examples, the light emitted by the luminescent substance and captured by the detector 16 can be fluorescence.

[0045] In some exemplary methods described herein for training a machine learning system to quantify a target nucleic acid present in a biological assay (e.g., performed in a reaction vessel using a nucleic acid amplification device 8), the nucleic acid amplification device 8 may be a nucleic acid amplification device of a specified type. For example, the nucleic acid amplification device 8 may include one or more specific features and / or may be a specific model of a nucleic acid amplification device from a specific manufacturer. In some such examples, the trained machine learning system produced by such methods may be customized for a nucleic acid amplification device of a specified type, which may enhance the accuracy of the trained machine learning system. A nucleic acid amplification device having any suitable configuration may be used. For example, the nucleic acid amplification device may include a support (e.g., a rotating support) configured to receive a reaction vessel rather than a block. In some such examples, the reaction vessel may be a capillary or a tube of a more conventional configuration. In some examples, the detector 16 of the nucleic acid amplification device may be positioned above the reaction vessel or at any suitable location. Therefore, the configuration of the nucleic acid amplification device described herein is not intended to be limiting, but rather to illustrate an example.

[0046] Figure 1 The exemplary system also includes a user device 20, which may include a processor 23 and a memory 22 for storing parameters representing one or more trained machine learning systems 25. In one exemplary method, the user device 20 receives a data set for each sample tested from the control unit 18. In some such exemplary methods, each data set includes data representing the amount of light received by the detector 16 at a specific time during an amplification cycle for a given sample. As described below with respect to Figure 3As further discussed, the user device 20 can be a device such as a computer workstation, a tablet computer, or other such user device co-located with the nucleic acid amplification device 8 in the user's laboratory. The nucleic acid amplification device 8 can be configured to transmit the data set from the control unit 18 to the user device 20, such as via any suitable wired connection (e.g., metal traces, optical fiber, Ethernet, etc.), a wireless connection (e.g., a personal area network, a local area network, a metropolitan area network, a wide area network, a cloud-based system, etc.), or a combination of both. For example, the user device 20 may include a communication unit that includes a network interface card such as an Ethernet card, an optical transceiver, a radio frequency transceiver, Interface card, WiFi TM A radio component, USB or any other type of device that can send and receive information to / from the nucleic acid amplification device 8.

[0047] In some exemplary methods, the processor 23 may be configured to apply the trained machine learning system 25 stored in the memory 22 to the data set and estimate the amount of the target organism present in the bioassay based on the data set. In some examples, the processor 23 may store the estimated amount of the target organism, such as in association with other data about the bioassay. The estimated amount of the target organism may be compared with a corresponding threshold in a limit test to determine whether the sample passes or fails the limit test. In some such exemplary methods, the threshold may be a value associated with one or more regulatory standards, industry practices, or associated intervention processes. For example, the estimated amount of the target organism in the sample may help enable the evaluation of the effectiveness of an intervention program designed to improve process efficiency and / or reduce pathogen levels in food products, feed products, water, and / or corresponding preparation environments.

[0048] Like this, comprise the system and method that the machine learning system that is trained is applied to the data set associated with the amplification sample of target nucleic acid to estimate the amount of target organism in sample can help solve the public health problem associated with pathogen.For example, because the system and method for nucleic acid quantification described herein provide faster value than the traditional method for pathogen quantification, so this type of system and method can make pathogen quantification easier to be obtained by food industry.This increased accessibility can be used by food industry for example to obtain more subtle understanding of pathogen existence compared with simply obtaining by detecting the presence or absence of pathogen.Increased accessibility can also be used to support limit testing in pathogen analysis, because one goal of limit testing is to detect the concentration of foodborne pathogens that meets or exceeds the threshold concentration and limit the release of products that may adversely affect public health.

[0049] Figure 2 is a block diagram illustrating an exemplary system 6 according to one aspect of the present disclosure, the system comprising Figure 1A nucleic acid amplification device 8, an external device such as a server, a network, and an access point for coupling the nucleic acid amplification device to the external device via the network. Figure 2 As shown, system 6 may include access point 24, network 26, and one or more external devices, such as external device 28 (e.g., server), which may include processing circuitry 30 and / or memory 32. Figure 2 In the example shown, the nucleic acid amplification device 8 may use a communication circuit (not shown) for communicating via a wireless connection with the access point 24. The access point 24 then transmits information received from the nucleic acid amplification device 8 to the external device 28 via a wired connection through the network 26, and transmits information received from the external device 28 to the nucleic acid amplification device 8 via a wireless connection through the network 26.

[0050] The access point 24 may include a processor connected to the network 26 via any of a variety of connections, such as telephone dial-up, digital subscriber line (DSL) or cable modem, or other suitable connections. In other examples, the access point 24 may be coupled to the network 26 via different forms of connections, including wired or wireless connections. In some examples, the access point 24 may be a user device, such as a computer workstation or tablet computer that may be co-located with the nucleic acid amplification device 8 and the user. The nucleic acid amplification device 8 may be configured to transmit data to the access point 24, such as described above with respect to Figure 1 In addition, the access point 24 can query the nucleic acid amplification device 8, such as periodically or in response to a command from a user or from the network 26, to retrieve a data set related to one or more bioassays or retrieve other information stored in a memory (not shown) of the nucleic acid amplification device 8. The access point 24 can then transmit the retrieved data to the external device 28 via the network 26.

[0051] In some examples, memory 32 of external device 28 may be configured to provide a secure storage location for data collected from access point 24 and / or nucleic acid amplification device 8. In some examples, memory 32 stores parameters representing one or more trained machine learning systems 35. In some examples, external device 28 may aggregate the data in a web page or other document for a user to view via access point 24 or nucleic acid amplification device 8. Figure 2 One or more other computing devices of the system are viewed. In this way, Figure 2 The system can realize remote (e.g., cloud-based) storage and access of data associated with the user's testing of food or feed products and / or corresponding production environments. Such a system can be customized to meet the data storage and / or access needs of a specific user.

[0052] Figure 3 The present invention is a diagram showing an aspect of the present invention. Figure 1Schematic conceptual diagram of the features of user equipment 20. Although relative to Figure 1 The user device 20 describes Figure 3 , but one or more components of the user equipment 20 described herein may be similar in function and / or structure to Figure 2 24 and / or one or more components of the external device 28 shown in FIG. In one exemplary method, the user device 20 includes a user interface 40 and a computing device 42. The user interface 40 may include a display 38, a graphical user interface (GUI), a keyboard, a touch screen, a speaker, a microphone, etc.

[0053] Computing device 42 includes one or more processors 23, one or more input devices 46, one or more communication units 48, one or more output devices 50, and memory 22. In some examples, computing device 42 and user interface 40 are components of the same device, such as a computer workstation, a tablet computer, etc. In some such examples, user interface 40 may include one or more of input devices 46. In other examples, computing device 42 and user interface 40 are separate devices such that user interface 40 does not necessarily include one or more of input devices 46.

[0054] The one or more processors 23 of the computing device 42 are configured to implement functions, process instructions, or both for execution within the computing device 42. For example, the processor 23 is capable of processing instructions stored within the memory 22, such as instructions for applying a trained machine learning system to a data set to estimate an initial amount of a target nucleic acid or target organism present in a sample. Examples of the one or more processors 23 may include any one or more of a microprocessor, a controller, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or an equivalent discrete or integrated logic circuit.

[0055] In some examples, computing device 42 may utilize one or more communication units 48 to communicate with one or more external devices (e.g., Figure 2 The communication unit 48 may include a network interface card, such as an Ethernet card, an optical transceiver, a radio frequency transceiver, or any other type of device configured to send and receive information. The communication unit 48 may also include a WiFi TM Radio component or Universal Serial Bus (USB) interface.

[0056] In some examples, one or more output devices 50 of computing device 42 may be configured to provide output to a user using, for example, audio, video, or tactile media. For example, output device 50 may include display 38 of user interface 40, a sound card, a video graphics adapter card, or any other type of device for converting signals (such as signals associated with information about the status, results, or other aspects of one or more data sets of amplification cycles performed by nucleic acid amplification device 8 analyzed by a trained machine learning system) into an appropriate form understandable to humans or machines. In some exemplary methods, user interface 40 includes one or more of output devices 50 employed by computing device 42.

[0057] The memory 22 of the computing device 42 may be configured to store information within the computing device 42 during operation. In some examples, the memory 22 may include a computer-readable storage medium or a computer-readable storage device. The memory 22 may include temporary storage, which means that the main purpose of one or more components of the memory 22 may not necessarily be long-term storage. The memory 22 may include volatile memory, which means that the memory 22 does not retain the stored content when power is not provided to it. Examples of volatile memory include random access memory (RAM), dynamic random access memory (DRAM), static random access memory (SRAM), and other forms of volatile memory known in the art. In some examples, the memory 22 can be used to store program instructions for execution by the processor 23, such as instructions for applying a trained machine learning system to a data set received from the nucleic acid amplification device 8 via one or more communication units 48. In some examples, the memory 22 can be used by software or applications running on the computing device 42 to temporarily store information during program execution.

[0058] In some examples, the memory 22 may also include a signal processing module 52, a training module 54, and a detection module 56. In some such examples, the detection module 56 includes a machine learning system (such as the machine learning systems 25 and 35) that, when trained, estimates the concentration of a target organism in a sample. In one such exemplary method, the training module 54 receives a data set of assays with known cell concentrations collected by the nucleic acid amplification device 8 during one or more amplification cycles, and uses the data set to train the detection module 56 to estimate the concentration of a target organism in a sample.

[0059] In some examples, the memory 22 may include a non-volatile storage element. Examples of such non-volatile storage elements include magnetic hard disks, optical disks, floppy disks, flash memory, or electrically programmable memory (EPROM) or electrically erasable programmable (EEPROM) memory. In one such exemplary method, the signal processing module 52 may be configured to analyze data received from the nucleic acid amplification device 8, such as a data set of time series measurements of a sample captured by the detector 16 and including light emitted by a luminescent substance within the sample during an amplification cycle, and process the data to improve the quality of the sensor data.

[0060] The computing device 42 may also include Figure 3 For example, computing device 42 may include a power supply for providing power to components of computing device 42. Similarly, in each example of computing device 42, Figure 3 The components of computing device 42 shown may not be required.

[0061] Figure 4 is a flow chart showing exemplary points for pathogen testing before, during and / or after food or feed production according to one aspect of the present disclosure. Figure 4 As shown, food production environment 60 may include raw materials 62. Food production process 64 that processes raw materials 62 and produces final product 66 may occur within food production environment 60. In some examples, production process 64 may occur entirely within food production environment 60, and raw materials 62 may be Figure 4 The process shown begins by entering food production environment 60 from outside of food production environment 60. In some examples, food production environment 60 can be an environment where food or feed materials are harvested, such as a greenhouse or field where such materials are grown. In some examples, the sample from food production environment 60 can be a water sample from a water source within food production environment 60, such as a water source used for washing and / or cooking.

[0062] Raw materials 62 may acquire pathogens from outside of food production environment 60 and introduce such pathogens into food production environment 60 when or after raw materials 62 are introduced into food production environment 60. Therefore, to help reduce foodborne illnesses caused by pathogens, there is an increasing trend toward pathogen testing of raw materials (e.g., raw materials 62) and food production environments (e.g., food production environment 60). In addition, pathogen testing of raw materials 62 may help prevent pathogen contamination of final product 66 (or other final products) by identifying contamination before the raw materials enter food production environment 60, so that contaminated raw materials may be prevented from entering food production environment 60.

[0063] The final product 66 may be located in the environment 60 for a period of time before being shipped out of the environment 60, such as before, during and after packaging. The final product 66 may acquire pathogens from the food production environment 60 or other sources within the food production environment 60, such as pathogens introduced by the raw materials 62. However, as described above, traditional pathogen quantification methods can be significantly time-consuming, requiring one or more days to produce results, and molecular methods for pathogen quantification have not yet been widely used. In some cases, the time required for traditional pathogen quantification methods can limit the food processing rate. In addition, due to time requirements, such traditional methods only provide current pathogen assessments as long as the sample is obtained, which may not provide an accurate assessment of the current state of the material, environment or product. Therefore, at least due to the time advantage of the molecular methods for pathogen quantification described herein, pathogen detection of raw materials 62, food production environment 60 and / or final product 66 (e.g., as part of release testing), such as at the testing point 68, according to such methods, the latest assessment can be provided, which can ultimately help prevent the release of contaminated final products to the public.

[0064] Figure 5 is a flow chart illustrating an exemplary technique for estimating the amount of a target organism in a sample according to one aspect of the present disclosure. Figure 5 An exemplary method may use a nucleic acid amplification device such as Figure 1 and Figure 2 The nucleic acid amplification device 8 of the system is performed. Figure 1 As described above, the nucleic acid amplification device 8 may be any suitable type of nucleic acid amplification device and may be configured to perform any suitable nucleic acid amplification technique, such as LAMP or PCR. Figure 1 is described in the context of a system, but Figure 5 The exemplary techniques may be performed using any suitable nucleic acid amplification device and computing equipment. Figure 5 More specific aspects and examples of the techniques generally described in the accompanying drawings will be described below with respect to 9A to 9C and Fig.11 Give a description.

[0065] exist Figure 5 In the exemplary method of the present invention, the nucleic acid amplification device 8 amplifies the target nucleic acid (80) in the enriched sample in the reaction chamber 10. In some examples, the sample can be derived from the food production environment 60, the raw material 62 or the final product 66, as described above with respect to Figure 4As described. The nucleic acid extracted from the sample can be placed in a reaction container (e.g., a PCR tube) and a luminescent substance that emits light that is in a stoichiometric relationship with the target nucleic acid, which can be a DNA sequence associated with a target organism (e.g., a bacterial genus or bacterial species). In some examples, the sample can be an enriched sample of a sample derived from a food or feed raw material, a final product, water, or a production environment. For example, the sample placed in the reaction container can be an enriched sample from a culture derived from an initial sample. In some such examples, the estimated amount of the organism can be the estimated initial amount of the organism. In some examples, such reaction containers containing samples and luminescent substances can be collectively referred to as "biological assays" herein. The detector 16 of the nucleic acid amplification device captures a data set of a time series measurement sample including light emitted by the luminescent substance in one or more amplification cycles, and transmits the data set to the computing device 42 of the user device 20, the computing device of the access point 24, or any other suitable computing device (82).

[0066] In the example of user device 20, one or more of processor 23, signal processing module 52, and / or other components of computing device 42 may apply a trained machine learning system to a data set to estimate the amount of a target organism in a sample (84). In some examples, the data set may include one or more data subsets associated with one or more different portions or phases of an amplification cycle (such as one or more portions or phases before, during, and / or after a peak amplitude of light emitted during an amplification cycle). Including data subsets from such different portions or phases of an amplification cycle may contribute to the accuracy with which a trained machine learning system can estimate the amount of a target organism in a sample, as described below with respect to Fig.11 and Fig.12 Further described.

[0067] Figure 6 and Figure 7 is a conceptual diagram illustrating representative features of an exemplary nucleic acid amplification technique that can be used with the systems and methods described herein. Figure 6 Describes technical aspects of exemplary LAMP technology, such as to which such technical aspects may be relevant to reaching Figure 6 The degree to which the examples are relevant. Figure 7 Aspects of exemplary qPCR techniques that can be used with the systems and methods described herein are shown. Figure 7 Discuss technical aspects of exemplary qPCR techniques, such as how such technical aspects may be relevant to reaching Figure 7 However, it should be understood that the systems and methods described herein may be used with any suitable nucleic acid amplification techniques and devices and are not limited to use with respect to Figure 6 and Figure 7 The specific examples described.

[0068] LAMP uses strand-displacing Bst DNA polymerase and four to six primers to produce continuous DNA amplification at a constant temperature (i.e., under isothermal conditions). In the LAMP technique, amplification and detection of target nucleic acids can be completed in a single step by incubating a mixture of samples, primers, DNA polymerases with strand-displacing activity, and substrates at a constant temperature (about 60°C to 65°C). In some examples, LAMP can provide high amplification efficiency, with DNA amplified by 10 within 15 minutes to 60 minutes. 9 Times – 10 10 Due to its high specificity, the presence of the amplified product can indicate the presence of the target gene.

[0069] In LAMP, four different primers recognize six different regions in the template (i.e., target) DNA sequence, and two loop primers recognize two additional sites in the corresponding single-stranded loop region during LAMP. The four different primers that recognize the six different regions of the target DNA may include a forward inner primer (FIP), a forward outer primer (F3; aka FOP), a reverse inner primer (BIP) and a reverse outer primer (B3; aka BOP). The two loop primers include a forward loop primer (FLP) and a reverse loop primer (BLP). In contrast, PCR and qPCR each use a non-strand displacement Taq DNA polymerase and two corresponding primers (forward primer and reverse primer) to recognize two different regions. In addition, qPCR uses a probe that is specific to the third different region (e.g., a fluorescent emission molecular beacon probe, a fluorescent emission hydrolysis probe, a primer carrying a fluorescent emission probe element, or another suitable probe comprising a fluorescent portion).

[0070] Two kinds of loop primers FL and BL can be bound to additional sites and accelerate the reaction during LAMP.For example, a primer containing a sequence complementary to the single-stranded loop region (between B1 and B2 regions or between F1 and F2 regions) on the 5' end of the dumbbell-shaped structure formed during LAMP can provide a starting point for DNA synthesis to increase the number during the LAMP technique.For example, an amplification product comprising six loops (not shown) can be formed during LAMP.In an exemplary technique that does not use loop primers FL and BL, four of the six in such loops will not be used.By using loop primers, all single-stranded loops can be used as the starting point for DNA synthesis, thereby reducing the amplification time.For example, the time required for amplification with loop primers can be about one-third to about one-half of the time required for amplification in the example that does not use loop primers.In some examples, when using loop primers, amplification can be achieved within 30 minutes.

[0071] Figure 6Real-time detection of nucleic acid amplification during LAMP amplification cycles based on measurements of bioluminescent intensity over time according to one aspect of the present disclosure is shown. In an exemplary LAMP technique, isothermal DNA amplification releases pyrophosphate (PPi) as a byproduct. The byproduct PPi is then converted into adenosine triphosphate (ATP) by the enzyme ATP-sulfurylase in the presence of adenosine 5'-phosphosulfate. In one such exemplary method, a bioassay having a sample of a target nucleic acid to be analyzed may be adapted to contain luciferase and its substrate luciferin, which may be used as a luminescent substance in the exemplary systems and methods described herein. Since ATP is a cofactor for the reaction of luciferase with luciferin to produce bioluminescence, the conversion of PPi to ATP during the amplification cycles of the LAMP technique drives the emission of bioluminescence. This emission of bioluminescence may be detected by a detector (such as a nucleic acid amplification device) configured for LAMP. Figure 1 and Figure 2 The nucleic acid amplification device 8 is used to detect the bioluminescence and store the data representing the time series measurement results of the bioluminescence as a data set. Figure 6 The illustrated mechanism for generating light during the LAMP technique may provide one or more other benefits, such as enabling real-time detection of nucleic acid amplification occurring during a LAMP amplification cycle within a relatively short period of time, such as approximately 15 minutes.

[0072] The time series measurement results of the relative light units (RLU) emitted by the luminescent material (e.g., fluorescein) in the biological assay containing the target nucleic acid are shown in curve 90. The time series measurement results of the relative light units (RLU) emitted by the luminescent material (e.g., fluorescein) in the control without the target nucleic acid are shown in the baseline curve 92. As shown in curve 90, the exponential amplification of the target nucleic acid during the LAMP amplification cycle produces a bioluminescent signal with both a rapid increase in RLU and a rapid decrease in RLU. In such examples, the time to reach the peak RLU emission corresponds to the amount of the target organism. For example, a relatively large amount of target organisms can produce a shorter time to reach the peak RLU emission. Therefore, one or more aspects of curve 90 (such as the time to reach the peak or the amplitude) can be used to train a machine learning system to estimate the amount of the target organism in the sample.

[0073] In some examples, a data set used to train a machine learning system (such as a neural network) includes data captured as a set of time series measurement samples of bioluminescence captured throughout an amplification cycle. In one such example, luminescence measurements are taken approximately every 5 seconds, which can be accumulated for reporting purposes as measurements at 10 second, 15 second, 20 second, and / or 25 second intervals across the amplification cycle.

[0074] In some exemplary methods, the data set used to train a machine learning system (such as a neural network) includes time series measurements of bioluminescence taken throughout the nucleic acid amplification cycle. In other exemplary methods, the training data set includes measurements taken during one or more of the first stage 94 of the amplification cycle, the second stage 96 of the amplification cycle, and the third stage 98 of the amplification cycle. In some such examples, the machine learning system can be trained to estimate the amount of target organism present in the sample based on samples in each of the first data subset, the second data subset, and the third data subset, based on the data set of samples collected throughout the amplification cycle, or based only on samples in the second subset. In one such exemplary method, samples from the second subset include samples from the T max The samples obtained at T max is the time at which the maximum amplitude of the target nucleic acid is detected during the nucleic acid amplification cycle. Likewise, samples may be collected approximately every 5 seconds, which may be accumulated to measurements of approximately 10 seconds, 15 seconds, 20 seconds, and / or 25 seconds across the amplification cycle for reporting purposes. Training a machine learning system based in part on a subset of data not associated with peak amplification may provide more robust training than training based solely on one or more subsets of data associated with peak amplification, which in turn may enhance the ability of the trained machine learning system to accurately estimate unknown quantities of target organisms.

[0075] A detector (such as detector 16 of nucleic acid amplification device 8) may capture a data set including a time series measurement sample of light emitted by the luminescent substance during the amplification cycle, as shown by curve 90, and transmit the data set to a computing device (e.g., computing device 42), which may apply the trained machine learning system. Figure 6 The mechanism of light generation during the described LAMP technique may enable a user to obtain an estimated amount of a target organism in a sample more quickly than is possible using traditional pathogen quantification methods.

[0076] In PCR, DNA extension is limited to a specific period of each thermal cycle (i.e., amplification cycle). In PCR, the presence of inhibitors can prevent polymerase from extending DNA within the allowed time, which can result in incomplete amplification products and can prevent the detection of target organisms. The association and dissociation of polymerase with DNA template during the temperature cycle of PCR and the denaturation step provide many opportunities for inhibitors to interfere. Compared with systems based on PCR and immunoassays, inhibition may be less likely to occur in LAMP technology. In addition, PCR may be more likely to be interfered with by the natural fluorescence of some food samples and enrichment medium. Therefore, using LAMP technology in systems and methods described herein can provide one or more beneficial effects that are superior to using PCR technology. However, as mentioned above, in other examples, using PCR technology in conjunction with systems and methods described herein can provide one or more beneficial effects that are superior to traditional pathogen quantitative methods.

[0077] Figure 7 Detection of nucleic acid amplification during an exemplary qPCR technique across multiple PCR cycles based on measurements of fluorescence intensity over time according to one aspect of the present disclosure is shown. In some such examples, the luminescent substance may be a fluorescent emitting hydrolysis probe, such as a TaqMan hydrolysis probe (available from Thermo Fisher Scientific). During PCR, the 5'-3' exonuclease activity of the Taq polymerase cleaves the probe into two parts 100A and 100B during hybridization with a complementary target DNA sequence. Cleavage of the hydrolysis probe produces a fluorescent signal, which is detected during the PCR reaction. Figure 7 It is represented by curve 102.

[0078] As shown in curve 102, the amplification of the target nucleic acid during the PCR operation including multiple amplification cycles produces a fluorescent signal. Curve 102 may include several parts or stages of the corresponding parts or stages reflecting the amplification of the target nucleic acid. For example, curve 102 may include a first portion 104 corresponding to the initial stage of amplification, during which the fluorescent signal may remain below a threshold. Curve 102 may also include a second portion 106 corresponding to the exponential amplification stage, during which the fluorescence exceeds the threshold and increases exponentially. Finally, curve 102 may include a third portion 108 corresponding to the stable stage of amplification, during which the fluorescence remains above the threshold and slowly increases in additional amplification cycles.

[0079] and Figure 6As with the exemplary LAMP technique of the present invention, a machine learning system can be trained to estimate the amount of a target organism present in a sample based on each of a first data subset, a second data subset, and a third data subset corresponding to a respective stage in the first stage, the second stage, and the third stage of the fluorescence signal as described above. The machine learning system can also be trained to estimate the amount of a target organism present in a sample based on a data set of fluorescence signal measurements collected throughout the amplification cycle. Training a machine learning system based in part on a data subset that is not associated with the exponential amplification phase of a PCR run (e.g., background fluorescence generated at the beginning of an amplification cycle) can provide more robust training than based solely on one or more data subsets associated with peak amplification (e.g., a subset that includes at least the exponential phase), which in turn can enhance the ability of the trained machine learning system to accurately estimate an unknown amount of a target organism.

[0080] Figure 8 The relationship between the time to peak amplitude and cell count in the LAMP molecular assay of five Salmonella strains according to one aspect of the present disclosure is shown. In some exemplary methods, the quantification of DNA-based assays is performed using high-quality DNA and a single response value from a DNA amplification reporter gene. The response value (usually fluorescence or bioluminescence) can be based on a signal exceeding a preset threshold or based on a peak amplitude value. Figure 8 A linear model with responses from five Salmonella strains is shown, with n = 480. In some examples, it may be desirable to estimate the initial amount of more than one strain or species (e.g., within a genus) of a target organism in a sample, since more than one of such strains or species may be pathogenic. Methods using multiple strains of a target organism will be described in Figure 8 discussed in the context of .

[0081] exist Figure 8 In the example shown, culture preparation was performed by inoculating 10 mL of buffered peptone water (BPW, 3M Company, St. Paul) with a single colony from an agar plate corresponding to each strain (Table 1). The inoculated liquid culture medium was incubated at 37°C for 18 hours.

[0082] Table 1 .

[0083]

[0084] 1 American Type Culture Collection and Tecra TM Collection Center.

[0085] For enumeration, cultures were serially diluted in Butterfields buffer and plated on 3MTM BrandPetrifilm TM Aerobic Count (AC) plates (3M Company) (hereinafter referred to as "Petrifilm AC plates"). The culture was maintained at 4°C-8°C until the plate count results were obtained. 3M TM Brand Molecular Detection Assay 2 – Salmonella (3M Company) (hereafter “MDA2 – Sal”), the counts obtained are used to estimate the number of cells used for the assay. When performing the detection assay, Petrifilm AC plates are used for final plate counts. These final plate counts are used to report the cell concentrations.

[0086] In one exemplary method, each strain is serially diluted in Butterfield buffer to approximately 10 2 10 3 10 4 10 5 and 10 6 Colony forming units (CFU) / mL were measured. Aliquots from each dilution were analyzed using MDA2-Sal according to the manufacturer's instructions. The MDS software provided by 3M was then used to determine the peak time-response to target sequence amplification. Figure 8 The peak time response of each aliquot at the cell concentration of the aliquot determined from the final plate count of each strain is shown. The data set of peak times of cells of known concentrations was then used to train a decision forest regression model and a boosted decision tree model. Both methods produced a coefficient of determination of approximately 0.75. The same data set used to train a linear regression model around line 110 produced a coefficient of determination R of approximately 0.2912 2 Other regression techniques such as support vector regression, random forest regression, ridge regression, logistic regression, Lasso, and nearest neighbor regression can also be used to train a model based on a dataset of peak times of cells of known concentration.

[0087] The peak time response is not always the best measure of cell count. Different matrices (i.e., materials other than pure cultures in samples or molecular components in food samples) may hinder the good consistency between the peak time response and the actual cell count. The specific count of the cells of the Salmonella strain can, for example, produce different peak time measurements according to the matrix where the cells are located. For example, different peak time measurements can be caused by the specific count of the cells of the Salmonella strain in the shellfish matrix or other different matrices by the salmon matrix. In some exemplary methods, the measurement results of the parameters (such as light intensity) over time across the nucleic acid amplification cycle provide a better representation of the initial cell count. Even so, it may be advantageous to train a machine learning system with different matrices to more accurately estimate the amount of the target organism in a specific matrix.

[0088] 9A to 9C and Fig.10 Exemplary systems and techniques for training and using machine learning systems to quantify biological activity in biological assays, such as those comprising Figure 8 Bioassays of said Salmonella species. 9A to 9C is a flow chart illustrating an exemplary method for training a machine learning system and for employing the trained machine learning system to quantify a target organism of interest in a sample according to aspects of the present disclosure. Fig.10 It is shown that it can be used for training Fig. 9A and Fig. 9B A block diagram of a system used in an exemplary technique of a machine learning system. 9A to 9C The systems and methods for training and using machine learning systems described in the exemplary technology are similar to those described in the exemplary technology. Figure 8 The predictive power of the constructed model was improved compared to the model based on the time-to-peak measurement results shown in the model shown. In addition, such systems and methods for training and using machine learning systems perform well compared to traditional methods when there is a specific matrix involved (e.g., poultry rinse matrix) rather than just pure culture.

[0089] exist Fig. 9A In an exemplary method of the present invention, the nucleic acid amplification device 8 in the system 6 is used to test an assay of a target organism with a known cell concentration to obtain a data set for each assay (112). The assay may be from a culture, from a matrix, or both. Each data set is then labeled with an amount reflecting the amount of the target organism detected by the nucleic acid amplification device in each corresponding array (114). The system 6 then trains a machine learning system using the labeled data set (116). In some exemplary methods, the method also includes using the trained machine learning system to estimate the amount of the target organism in the assay (118). In some exemplary methods, an alternative quantitative method such as, for example, MPN is used to label each data set with an amount obtained from the corresponding assay.

[0090] In some exemplary methods, each data set includes a time series measurement sample of the light intensity detected by the detector 16 during the amplification cycle. Each data set is marked with the known cell concentration of its corresponding assay, and the marked data set is then used to train the machine learning system 25 or 35, as described in detail below. The machine learning system 25 or 35 is then used to estimate the amount of the target organism in each assay. In some exemplary methods, different data sets are used for each matrix or matrix type. The matrix representing the target organism in cheese can be used for example to train the machine learning system 25 or 35 for quantitative target organisms in a cheese factory.

[0091] Fig. 9B is another flow chart illustrating an exemplary method for obtaining a data set from a matrix having a known cell concentration and for using the data set to train a machine learning system to quantify a target organism of interest in the matrix. Fig. 9B In an exemplary method of the invention, the method includes obtaining a sample from a matrix to be tested (122), adding an enrichment medium to the sample (124), diluting the sample (126), and then incubating the sample (128), after which the sample is analyzed with a nucleic acid amplification device to generate a data set (130). The method also includes testing the sample using an alternative method (such as MPN) to generate a label for each data set having a known cell concentration of the sample that generated the data set (120). The data set and its associated label are then used to train a machine learning system (132).

[0092] In some exemplary methods, each data set includes light intensity measurements taken over time during one or more amplification cycles. In some such exemplary methods, each data set includes time series measurements of light intensity captured across an entire amplification cycle. In some exemplary methods, such data sets also include measurements taken during a period at the beginning of an amplification cycle, where data is typically not captured, discarded, or otherwise suppressed by the nucleic acid amplification device 8. In some exemplary methods, each data set includes a time series of measurements of light intensity captured across an entire amplification cycle. max The light intensity measurement results taken in the first period before, during the period including T max The light intensity measurements taken during the second time period and during T max The light intensity measurements are then taken during the third time period.

[0093] exist Fig. 9B and Fig. 9C In the exemplary method of , steps 120-132 can be performed in an exemplary technique for training a machine learning system, and steps 122-130 and 136 can be performed in an exemplary technique for using a trained machine learning system. Although one or more aspects of the two workflow techniques may be described herein with respect to one or more specific nucleic acid amplification and detection components, in other examples, Fig. 9B and Fig. 9C The techniques may be performed using one or more other nucleic acid amplification and detection components.

[0094] In one such exemplary method of using a machine learning system to estimate the amount of a target organism, Fig. 9C The technique includes receiving a sample of a matrix (122), such as by a laboratory worker or an automated device. The matrix can be, for example, a matrix in which a target organism may be present, such as a Figure 8The poultry rinse matrix or a portion of the raw material of the food product or the final product of the food product. Upon receiving the matrix, laboratory personnel or equipment adds an appropriate enrichment medium that is configured to grow the target organism within the sample containing the target organism and the matrix to detectable limits (124). In some examples, such as examples where PCR techniques are used to amplify the target nucleic acid, the appropriate enrichment medium may have a property that is less likely to interfere with the fluorescence emitted during PCR than one or more other appropriate enrichment media (such as by emitting less background fluorescence relative to other appropriate media). Next, in some exemplary methods, the personnel or equipment prepares a 1:10 dilution of the resulting enrichment solution (126). As described below with respect to Fig.11 and Fig.12 As described above, using a 1:10 dilution can increase the specificity of the trained machine learning system for the target organism. Any other suitable dilution can be used, such as 1:100 or 1:1000. In some exemplary methods, the amount of dilution will depend on system characteristics, such as the type of targeted organism and the specific amplification technology.

[0095] Next, the sample within the enrichment solution is incubated to allow enrichment of the target organism (128). In some examples, the sample may be incubated at about 35°C-42°C for about 4 hours-24 hours or at any other suitable temperature and time period that allows the target organism to grow appropriately. In other examples, an enrichment step may not be used, and nucleic acids may be extracted from the sample without enrichment. After incubation, if used, in some exemplary methods, the sample is analyzed via amplification and detection of target nucleic acids associated with the target organism (130). For example, a nucleic acid amplification device 8 having a light detector 16 (such as an MDS) can be used to amplify and detect target nucleic acids. For example, the MDS can be configured to amplify target nucleic acids by performing a LAMP technique, and then the detector 16 can be used to detect bioluminescence emitted by a luminescent substance (e.g., luciferin) within the sample. By combining LAMP with bioluminescent detection, a nucleic acid amplification device (such as an MDS) can make molecular detection of foodborne pathogens simpler and faster, thereby providing users with the speed and convenience of simultaneously identifying one or more target organisms (e.g., one or more species or strains of Salmonella, Listeria, Listeria monocytogenes, Escherichia coli O157 (including H7), Campylobacter, Cronobacter, and / or other target organisms) in food and / or environmental samples. In other exemplary methods, different LAMP platforms are used or different nucleic acid amplification platforms are used to perform the PCR. 9A to 9C technology.

[0096] In some exemplary methods, the amplitude of light generated early in the amplification cycle (e.g., before stage 94 or stage 104) may be suppressed (e.g., not recorded) so as not to confuse the user with background activity. However, it has been found that such information can be helpful in training a machine learning system. Thus, in one exemplary method, the data set includes Figure 6 The time series measurements taken before stage 94 in the example method. In a similar exemplary method, the data set includes Figure 7 The time series measurement results performed before stage 104 in FIG.

[0097] In some exemplary methods, the labeled data sets are generated by an expert examining individual samples on which nucleic acid amplification has been performed. In one such exemplary method, the expert receives the data sets associated with the samples, determines the amount of the organism and / or target nucleic acid in the sample (via, for example, one of the conventional quantification techniques described above, such as MPN), and labels each data set with the determined numerical value. The labeled data sets are then used to train a machine learning system, such as Fig. 9A and Fig. 9B shown.

[0098] In some exemplary methods, the data set includes time series measurements taken at predetermined intervals (e.g., 25 seconds) over the entire amplification cycle. In other exemplary methods, the data set includes data selected from certain stages of the amplification cycle. For example, the data set may include data from Figure 6 One or more of stages 94, 96 and 98 or from Figure 7 For example, where (130) includes LAMP technology, the data set may include one or more data subsets, such as Figure 6 For example, a data set may include: a first data subset representing a time series measurement sample of light emitted in an amplification cycle up to a first time point, the first time point occurring before a peak amplitude of the light emitted in the amplification cycle; a second data subset representing a time series measurement sample of light emitted after the first time point but before a second time point in the amplification cycle, the second time point occurring after the peak amplitude; and a third data subset representing a time series measurement sample of light emitted after the second time point in the amplification cycle. A computing device (e.g., Figure 2The computing device may train a machine learning system to predict an initial concentration (i.e., amount) of a target organism of interest using processing circuitry 30 of an external device 28 or any other suitable computing device (132). For example, the computing device may label a data set and / or one or more subsets of a data set with an estimate of the amount of the target organism within a bioassay associated with the corresponding data set or data subset. The computing device may then train the machine learning system with the labeled data set (or data subset) and / or substrate identification to estimate the amount of the target organism within the sample, thereby obtaining a trained model. The computing device may then store the parameters of the trained machine learning system in one or more storage components of the system, such as a memory of the computing device, a memory of a computing device of a user device 20, a memory of a computing device of an access point 24, and / or any other suitable location.

[0099] In a workflow technique associated with using a trained machine learning system to calculate quantities of an organism of interest, Fig. 9C The technique includes performing steps 122-130 substantially as described above with respect to the exemplary technique for training a machine learning system, although the substrate at (122) may be a sample of a raw food material, a final food product, or an environmental sample that may contain a target organism of interest rather than a known amount of the target organism. In such examples, a nucleic acid amplification and detection system (such as an MDS) or another system configured to perform LAMP or PCR during one or more amplification cycles and detect light emitted by a luminescent substance may capture a data set that includes a time series measurement sample of light emitted by the luminescent substance during the amplification cycle, and analyze the data set (130). The data set is then analyzed based on the trained machine learning model to obtain an estimate of the amount of the target organism in the substrate (136).

[0100] In some such examples, the data set may include one or more data subsets corresponding to one or more portions of an amplification cycle, such as in a manner similar to the data subsets with which the machine learning system is trained. For example, a data set corresponding to a sample containing an unknown amount of a target organism may include: a first data subset representing a time series measurement sample of light emitted in the amplification cycle until a first time point, the first time point occurring before the peak amplitude of the light emitted within the amplification cycle; a second data subset representing a time series measurement sample of light emitted after the first time point but before a second time point in the amplification cycle, the second time point occurring after the peak amplitude; and a third data subset representing a time series measurement sample of light emitted after the second time point in the amplification cycle. A computing device (e.g., computing device 42 of user device 20, computing device of access point 24, or any other suitable computing device) configured to receive the first data subset, the second data subset, and the third data subset applies the trained machine learning system to the data subsets (136) and calculates the concentration (e.g., amount) of the target organism of interest in the sample. In some examples, the computing device may then store one or more such estimates to one or more storage components of the system, such as memory of the MDS, memory of computing device user device 20, memory of a computing device of access point 24, and / or any other suitable location.

[0101] In some exemplary methods, separate machine learning systems are trained based on the type of substrate being tested. For example, separate systems may be trained for testing cheese or for testing feed, where parameters for each machine learning system are stored in memory based on the type of substrate being tested.

[0102] Fig.10 is a block diagram illustrating a device training system according to one aspect of the present disclosure. Fig.10 In the example shown, the device training system 140 includes a training module 144 connected to a labeled dataset module 146 via a link 148. The training module 144 is also connected to a machine learning system memory 150 via a link 152. In some exemplary approaches, the device training system 140 is connected to a user device 156 via the link 154. In one exemplary approach, the training module 144 includes a computing device, one or more storage components, and a user interface. For example, the device training system 140 may include Figure 2The computing device and memory 32 of the external device 28 of the training module 144. In one exemplary method, the training module 144 receives the labeled data sets from the labeled data set module 146. In some such exemplary methods, each labeled data set includes the amount of target organisms associated with the sample and the measurement results of light detected by the nucleic acid amplification device 8 during the amplification cycle of the sample. The training module 144 trains the machine learning system with the labeled data sets and stores the parameters associated with the machine learning system in the algorithm 150.

[0103] Obtaining labeled data can be time consuming because the generation of labeled data requires an expert to examine each sample or to generate a reference sample that can be used for comparison with the sample being measured. Alternatively, in the absence of labeled data, labeled data can be approximated by carefully controlling the environment in which the sample is collected. An exemplary method for generating labeled data from a reference sample is discussed below.

[0104] For enumeration, cultures were serially diluted in Butterfields buffer and plated on 3M TM BrandPetrifilm TM Aerobic Count (AC) plates (3M Company) (hereinafter referred to as "Petrifilm AC plates"). The culture was maintained at 4°C-8°C until the plate count results were obtained. 3M TM Brand Molecular Detection Assay 2 – Salmonella (3M Company) (hereafter “MDA2 – Sal”), the counts obtained are used to estimate the number of cells used for the assay. When performing the assay, Petrifilm AC plates are used for final plate counts. These final plate counts are used to report cell concentrations. In one example method, each strain is serially diluted in Butterfield buffer to approximately 10 2 CFU / mL, 10 3 CFU / mL, 10 4 CFU / mL, 10 5 CFU / mL and 10 6 CFU / mL. An aliquot from each dilution was analyzed using MDA2-Sal according to the manufacturer's instructions. The MDS software provided by 3M was then used to determine the peak time-response to target sequence amplification.

[0105] Figures 11 to 14 Techniques for using a trained machine learning system to predict the amount of five Salmonella species in a sample poultry rinse are shown. Fig.11Techniques for training a machine learning model to estimate the cell count of target cells seeded into a matrix, and techniques for using the trained machine learning model to estimate the cell count in the matrix based on the trained model, according to one aspect of the present disclosure are shown. Fig.11 In an exemplary method of the technology, poultry rinse (200) is prepared by adding 400 mL of BPW to the whole poultry carcass and mixing by hand. After removing the carcass, approximately 10% of each strain in Table 1 above is used. 1 10 2 10 3 and 10 4 10 mL aliquot (202) of the wash solution is inoculated with cells / sample. Figure 8 The strains are prepared as described in the examples used in the discussion of . In one exemplary method, enrichment medium is added to each aliquot (204). Fig.12 In the method shown, the matrix is ​​not diluted at 206, and Fig.13 In the example shown, the enrichment matrix is ​​diluted in a 1:10 dilution (206). The inoculated wash solution is then incubated at 41.5°C for 7 hours (208). After incubation, aliquots from the wash solution are analyzed using MDA2-Sal according to the manufacturer's instructions. The signal response (relative light units) of each aliquot is captured as a series of measurements taken over approximately 60 minutes (i.e., within the DNA amplification cycle of the MDS), and the data representing the measurements is stored in a data set associated with each aliquot (210).

[0106] In some exemplary methods, each data set includes a first data subset, a second data subset, and a third data subset. The first data subset includes measurements captured before a first time point in the amplification cycle, the first time point occurring before time Tmax, wherein time Tmax corresponds to the time at which the parameter measured in the nucleic acid amplification cycle reaches a peak amplitude. The second data subset includes measurements captured after the first time point in the nucleic acid amplification cycle but before a second time point, the second time point occurring after Tmax. The third data subset includes measurements captured after the second time point in the nucleic acid amplification cycle.

[0107] In training mode, each data set is labeled with a cell concentration based on an estimate of the initial cell concentration in the aliquot associated with the data set. In other exemplary methods, each data set is labeled with a value obtained via another method (such as MPN). A machine learning model such as a neural network is then trained using the labeled data sets to estimate the cell concentration in the matrix (212).

[0108] In production mode, the machine learning system receives a dataset for each matrix analyzed by the nucleic acid detector and determines an initial concentration of the target organism in the matrix by applying the dataset to the trained machine learning model (214). Fig.12 An example showing the difference between the predicted cell concentration from the neural network-based machine learning model and the cell concentration determined from the corresponding plate counts is shown in FIG. Fig.12 In the example shown, the model is able to explain 84% of the overall variability in the dataset.

[0109] In one exemplary method, Fig.11 The technique is performed at each known level inoculation of the organism of interest. In one such exemplary method, the process is repeated enough times to establish a representative sample of a data set at each CFU inoculation level in multiple CFU inoculation levels. In some such exemplary methods, for each type of matrix, this may require running 100 or more amplification cycles at each inoculation level. In one such exemplary method, the level includes a level below 10CFU, such as 1CFU-10CFU, a level between 10CFU-100CFU, a level between 10CFU-1000CFU, a level above 1000CFU and / or any other suitable known inoculation level. For each known level of inoculation, the nucleic acid amplification and detection device can capture a data set including a time series measurement sample of the light emitted by the luminescent substance during each amplification cycle.

[0110] Fig.13 The logarithmic difference between the cell count predictions made by a trained machine learning system according to one aspect of the present disclosure and the different cell counts of Salmonella cells inoculated into poultry wash matrix and a 1:10 dilution of the poultry wash matrix is ​​shown. Fig.12 An approach similar to the one used in the example of Fig.13 In the example, a 1:10 dilution of the flushing solution ( Fig.11 206) were also incubated and included in the analysis.

[0111] In one such exemplary method, poultry rinse was prepared by adding 400 mL of BPW to a whole poultry carcass and mixing by hand. After removing the carcass, approximately 10% of each strain in Table 1 above was added. 1 10 2 and 10 3 Inoculate 10 mL aliquots of the wash solution with 10 cells / sample. Figure 8For each flush, a 1:10 dilution was also prepared in BPW. The inoculated flush and dilutions were incubated at 41.5°C for 7 hours. After incubation, aliquots from all samples were analyzed using MDA2-Sal according to the manufacturer's instructions. Fig.11 and Fig.12 As in the example discussed, during DNA amplification, the entire signal response (relative light units) over time (60 minutes) was extracted, labeled, and used to train a neural network algorithm. 0 Dilution and 10 1 The response data for the dilutions were processed as a single data point and labeled with the concentration of cells seeded into the wash. For this exemplary method, Fig.13 The difference between the predicted cell concentration from the neural network model and the cell concentration from the corresponding plate counts is shown in .

[0112] In this case, the model was able to explain 99% of the overall variability in the dataset, outperforming Figure 8 The linear model is significantly improved and outperforms Fig.12 Improvements of the exemplary method. Fig.13 The results shown suggest that in some examples it may be desirable to include such dilutions when performing techniques for training a machine learning system.

[0113] Fig.14 Various metrics for measuring the performance of regression for cell count prediction using various machine learning techniques in accordance with aspects of the present disclosure are shown. Fig.14 In the example shown, the above Fig.13 Poultry rinses were prepared and tested as described in the exemplary methods of Fig.13 In the example shown, the 0 Dilution and 10 1 The response data of both dilutions were treated as a single data point and labeled with the cell concentration inoculated into the rinse. The labeled data set was then used to train a neural network model, a linear regression model, a Bayesian linear regression model, a decision forest regression model, and an enhanced decision tree regression model. Each model was used to predict the cell concentration. Fig.14 Metrics comparing the results from each machine learning model to traditional plate counts are provided.

[0114] Thus, as described herein, it may be advantageous to apply a trained machine learning system to a dataset of nucleic acid amplification bioassays derived from nucleic acids associated with one or more target organisms. Figure 8Compared to the model shown in 9A to 9C and Fig.11 For example, by applying a trained machine learning system to Figure 8 For a dataset with a 100% confidence interval, such methods use decision forest regression or boosted decision trees to obtain an R of 0.75. 2 .therefore, Fig.12 and Fig.13 It is shown that in addition to reducing or eliminating the need to isolate pure DNA for pathogen quantification, the systems and methods described herein can also perform well on multiple strains or species of organisms of interest, such as multiple Salmonella species.

[0115] As described above, assays based on molecular methods such as nucleic acid amplification (e.g., LAMP or PCR) can be affected by the presence of matrix-derived substances that can interfere with or prevent the reaction from proceeding correctly. In food production, matrix-derived substances (such as spices and environmental samples) can act as inhibitors that can interfere with nucleotide amplification assays (such as PCR and LAMP), resulting in false negative results or positive detections with incorrect quantification.

[0116] It may be difficult to eliminate inhibition or limit its effect. Careful sample handling can be used, for example, to remove inhibitory substances. However, sample handling cannot be relied upon to completely remove inhibitory substances. Inhibition can be detected via amplification controls; such controls can be used, for example, to verify that an assay has been performed correctly. Amplification controls add expense and complexity to molecular methods.

[0117] Fig.15 is a conceptual diagram illustrating nucleic acid amplification in a standard sample and an inhibited sample during a LAMP amplification cycle according to one aspect of the present disclosure. As described above, in LAMP, the emission of bioluminescence can be detected by a detector (such as a nucleic acid amplification device) configured for LAMP. Figure 1 and Figure 2 The data representing the time series measurement results of the bioluminescence intensity are stored as a data set. In some examples, the data are used to Fig.15 The illustrated mechanism for generating light during the LAMP technique may provide one or more other benefits, such as enabling real-time detection of nucleic acid amplification occurring during a LAMP amplification cycle within a relatively short period of time, such as approximately 15 minutes.

[0118] Inhibition can be manifested in several ways. Time to peak is a characteristic observed when evaluating a reaction for inhibition or other problems (due to poor reaction performance of primer design). Fig.15In the example, samples show a "normal" run (300, 302) and a post-peak (304, 306) of a run in which the matrix is ​​known to cause inhibition. Inhibited samples may tend to exhibit a longer time to peak RLU emission and a lower maximum amplitude. Similarly, in PCR, the presence of an inhibitor may prevent the polymerase from extending DNA within the allowed time, which may result in incomplete amplification products and may prevent detection of target organisms.

[0119] However, the difference in peak time can also be a response to different DNA concentrations. Therefore, it may be difficult to determine whether the shift in peak is a product of DNA concentration or due to some inhibition. Fig.16 The method described in the context of identifies and corrects quantification due to inhibition by training a machine learning system with a dataset from assays with different inhibition levels.

[0120] Fig.16 is a flow chart illustrating an exemplary technique for training a machine learning system to quantify a target organism in an inhibited sample according to one aspect of the present disclosure. The method can be used, for example, in a biological assay (such as one including Figure 8 Quantification of organisms in a bioassay of Salmonella species described herein. Figure 8 Systems and methods based on this approach improve the predictive power of the constructed model compared to the model based on the time-to-peak measurement results shown in the model shown. In addition, such systems and methods for training and using machine learning systems perform well in the presence of specific matrices involved (e.g., poultry rinse matrix) rather than just pure cultures, even in the face of inhibitory substances, compared to traditional methods.

[0121] exist Fig.16 In the example shown, a machine learning system (such as Figure 1 and Figure 2 Machine learning systems 25 and 35) are used to quantify the target organisms present in the biological assay. In one exemplary method, such as Fig.10 The illustrated device training system 140 receives a plurality of data sets, each data set being associated with a biological assay to be tested against a target organism (310). The plurality of biological assays include an inhibitory substance.

[0122] In one exemplary method, each data set includes data collected by the detector across one or more nucleic acid amplification cycles. The data includes activity measurements taken at different times during one or more nucleic acid amplification cycles, and represents nucleic acid amplification of target nucleic acids associated with target organisms within biological assays. In some exemplary methods, the activity measurements include time series measurements of relative light units (RLU) emitted by a luminescent substance (e.g., fluorescein) in a biological assay containing target nucleic acids. As described above, the exponential amplification of target nucleic acids during LAMP amplification cycles produces bioluminescent signals with both rapid increases in RLU and rapid decreases in RLU. In such examples, the curve tracked by the measurement results of RLU emission corresponds to the amount of target organisms present in the assay, even in the face of inhibition. Therefore, the device training system 140 can use parameters representing the curves tracked during one or more amplification cycles to train a machine learning system to estimate the amount of target organisms in a sample. Relevant parameters may include time to peak, but as described above, time to peak response is not always the best measure of cell counts. Measurements of parameters (such as light intensity) over time across nucleic acid amplification cycles provide a better representation of initial cell counts. In some exemplary methods, the measurements of light intensity over time include intensity measurements taken during amplification cycles but before amplification of the target nucleic acid is detected. Even so, it may be advantageous to train a machine learning system with different matrices and different levels of inhibition to more accurately estimate the amount of target organisms within a particular matrix.

[0123] In some examples, a data set used to train a machine learning system (such as a neural network) includes data captured as a set of time series measurement samples of bioluminescence captured throughout an amplification cycle, for both standard and suppressed bioassays. In one such LAMP example, luminescence measurements are taken approximately every 5 seconds, which can be accumulated as measurements across amplification cycles at intervals of 10 seconds, 15 seconds, 20 seconds, and / or 25 seconds for reporting purposes.

[0124] Back to Fig.16As discussed above, each data set received by the device training system 140 is labeled with an estimate of the amount of the target organism present in the associated bioassay (312). The labeled data sets are then used to train a machine learning system to estimate the amount of the target organism in the selected bioassay (314). In one exemplary method, the machine learning system is trained based on the activity measurements stored in each of the plurality of data sets and the estimate of the amount of the target organism present in the bioassay associated with each respective data set. In one exemplary method, the labeled data sets are used to train a model, such as a neural network model, a linear regression model, a Bayesian linear regression model, a decision forest regression model, and a boosted decision tree regression model, as described above in Fig.14 discussed in the context of .

[0125] exist Fig.16 In the exemplary method of the present invention, the nucleic acid amplification device 8 in the system 6 is used to test the assay of the target organism with known cell concentration, including the inhibition assay, to obtain the data set of each assay. The assay can be from culture, from matrix or both. Each data set is then marked with the amount reflecting the amount of the target organism detected by the nucleic acid amplification device in each corresponding array (312). Then the system 140 uses the marked data set to train the machine learning system (314). In some exemplary methods, the method also includes using a trained machine learning system to estimate the amount of the target organism in the assay (316). In some exemplary methods, an alternative quantitative method such as MPN is used to mark each data set with the amount obtained from the corresponding assay. In some exemplary methods, different data sets are used for each matrix or matrix type. The matrix representing the target organism in cheese can be used for example to train the machine learning system 25 or 35 for quantitative target organisms in a cheese factory.

[0126] In one exemplary method, a system for quantifying the presence of a target organism in a sample includes: a detection device (e.g., Figure 1 and Figure 2 A nucleic acid amplification device 8 in the embodiment of the present invention is configured to amplify and detect a target nucleic acid associated with a target organism, and a machine learning system (such as Figure 1 Machine learning systems in 25 or Figure 2 The machine learning system 35 in the sample is configured to receive the activity measurement results and estimate the amount of the target organism in the sample based on the activity measurement results. The detection device includes a detector and a reaction chamber, which is configured to receive the assay of the sample and amplify the target nucleic acid in the assay within a nucleic acid amplification cycle. The detector is configured to capture the activity measurement results representing the amount of target nucleic acid present in the assay at different times within the nucleic acid amplification cycle.

[0127] In one such exemplary method, a machine learning system is trained with a plurality of training data sets, each training data set being associated with a training assay and comprising activity measurements representing the amount of a target nucleic acid present in the training assay, wherein the training is based on the activity measurements stored in each training data set and an estimate of the amount of a target organism present in the training assay associated with each respective training data set. The training assays include assays having different inhibition levels.

[0128] As part of food, feed and water production safety, quantification of pathogens is becoming increasingly important. For example, for certain pathogens, such as B. cereus, S. aureus and Vibrio species, producers may be required to not only detect the presence or absence of the pathogen, but may be required to provide quantitative information about the pathogen. In addition, regulations in certain countries may require quantitative information for risk assessment; presence / absence criteria alone may not be sufficient to provide the required information. For example, in Europe, the maximum permitted level of Listeria monocytogenes in certain products varies depending on the intended use of the product.

[0129] Even in the case where regulations do not require, methods for obtaining quantitative pathogen information about pathogens can also be used to develop more effective intervention processes and / or more effective processes for monitoring pathogen levels than can be achieved using presence / absence standards. Food, feed and water producers may, for example, be able to use such methods to evaluate the effectiveness of current intervention programs in reducing pathogen levels in their products. Therefore, the ability to determine not only the presence of microorganisms present in biological assays but also the amount of microorganisms is becoming increasingly important not only in terms of quantitative pathogens, but also in terms of evaluating the efficacy of the steps taken to control pathogens in food, feed, water and corresponding processing environments. The ability to measure the amount of target organisms in the presence of inhibitors is particularly important. The above-mentioned technology provides rapid and accurate quantification of pathogens in samples, and can eliminate the need for amplification controls. In addition, since each type of microorganism is associated with one or more nucleic acids, the above-mentioned technology can be used to measure the cell concentration in samples containing any type of microorganisms.

[0130] Various examples have been described. These and other examples are within the scope of the following claims.

Claims

1. A system for quantifying a target organism present in a sample, the system comprising: A detection device, the detection device being configured to amplify and detect a target nucleic acid associated with the target organism, the detection device comprising: a reaction chamber configured to receive an assay of the sample and amplify the target nucleic acid in the assay in a nucleic acid amplification cycle; and a detector configured to capture activity measurements representative of the amount of the target nucleic acid present in the assay at different times within the nucleic acid amplification cycle and to store the activity measurements in a data set, wherein the data set comprises: A first data subset, the first data subset includes measurement results obtained up to a first time point, the first time point being at time T max Before, the time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; A second data subset includes measurements taken after the first time point but before a second time point in the nucleic acid amplification cycle, the second time point occurring at T max After that; and a third data subset comprising measurements taken after the second time point in the nucleic acid amplification cycle; and A machine learning system configured to receive the first, second, and third data subsets and to quantify the target organism in the sample based on the data subsets, wherein the machine learning system is trained to estimate the amount of the target organism present in the assay based on measurements present in the first, second, and third data subsets.

2. The system of claim 1, wherein the reaction chamber is configured to perform an amplification technique comprising one or more of LAMP, PCR, nucleic acid sequence-based amplification, or transcription-mediated amplification.

3. The system of claim 1, wherein the target organism is a microorganism of one or more Salmonella species, one or more Listeria species, one or more Campylobacter species, one or more Cronobacter species, one or more Escherichia coli strains, one or more Vibrio species, one or more Shigella species, one or more Legionella species, one or more Bacillus cereus strains, or one or more Staphylococcus aureus strains, one or more types of viruses, or one or more genetically modified organisms.

4. The system of claim 1 , wherein the reaction chamber is further configured to amplify the target nucleic acid in the sample within a plurality of nucleic acid amplification cycles, and Wherein the detector is further configured to capture the measurement results across the multiple nucleic acid amplification cycles.

5. The system of claim 1, wherein the machine learning system is based on a regression model.

6. The system of claim 1, wherein the reaction chamber is further configured to receive a module, wherein the module comprises: A plurality of first reaction containers, each container of the plurality of first reaction containers containing a certain amount of lysis buffer solution; as well as A plurality of second reaction containers, each of the plurality of second reaction containers containing a quantity of one or more reagents configured for a nucleic acid amplification reaction.

7. A method for quantifying a target organism present in a sample, the method comprising: receiving a plurality of data sets, wherein each data set is associated with a biological assay, each data set comprising measurements of a target nucleic acid detected within the associated biological assay, the measurements being performed by a nucleic acid amplification device of a specified type on the associated biological assay and collected over at least a portion of a nucleic acid amplification cycle, wherein the target nucleic acid is associated with a target organism; labeling each data set with an estimate of the amount of the target organism present within the associated biological assay; as well as training a machine learning system with the labeled data set to estimate the amount of the target organism within the biological assay based on testing of the target nucleic acid in the biological assay by the specified type of nucleic acid amplification device; wherein the measurements are time series measurements of light intensity collected over at least a portion of the nucleic acid amplification cycle, wherein each data set comprises: A first data subset, the first data subset includes measurement results obtained up to a first time point, the first time point being at time T max Before, the time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; A second data subset includes measurements taken after the first time point but before a second time point in the nucleic acid amplification cycle, the second time point occurring at T max after; as well as A third data subset includes measurements taken after the second time point in the nucleic acid amplification cycle.

8. The method of claim 7, wherein the nucleic acid amplification device performs an amplification technique comprising one or more of LAMP, PCR, nicking enzyme amplification reaction (NEAR), helicase-dependent amplification (HDA), nucleic acid sequence-based amplification (NASBA), or transcription-mediated amplification (TMA).

9. The method of claim 7, wherein the biological assay is from a matrix inoculated with two or more levels of an organism, and wherein labeling each data set with an estimate of the amount of the target organism comprises setting the amount as a function of the inoculation level.

10. The method of claim 7, wherein the biological assays are from a plurality of matrix types, and wherein training the machine learning system comprises training a machine learning model to distinguish between matrix types.

11. A non-transitory computer-readable medium storing instructions that, when executed by a processing circuit, cause the processing circuit to: receiving a data set generated by amplifying an amount of nucleic acid in a sample within a nucleic acid amplification cycle, wherein the nucleic acid is associated with a target organism, the data set comprising measurements collected during the nucleic acid amplification cycle representing the amount of nucleic acid in the sample, wherein the data set comprises: A first data subset, the first data subset includes measurement results obtained up to a first time point, the first time point being at time T max Before, the time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; A second data subset includes measurements taken after the first time point but before a second time point in the nucleic acid amplification cycle, the second time point occurring at T max after; as well as a third data subset comprising measurements taken after the second time point in the nucleic acid amplification cycle; as well as A machine learning system is applied to the data subset, wherein the machine learning system is trained to estimate an amount of the target organism present in the sample based on measurements present in the first data subset, the second data subset, and the third data subset.

12. The computer-readable medium of claim 11, wherein the measurements are time series measurements of light intensity collected within the nucleic acid amplification cycle.

13. A system for quantifying the presence of a target organism in a sample, the system comprising: A detection device, the detection device being configured to amplify and detect a target nucleic acid associated with the target organism, the detection device comprising: a reaction chamber configured to receive an assay of the sample and amplify the target nucleic acid in the assay in a nucleic acid amplification cycle; and a detector configured to capture activity measurements representative of the amount of the target nucleic acid present in the assay at different times within the nucleic acid amplification cycle; and a machine learning system configured to receive the activity measurements and to estimate the amount of the target organism in the sample based on the activity measurements, the machine learning system being trained with a plurality of training data sets, each training data set being associated with a training assay and comprising activity measurements representative of the amount of the target nucleic acid present in the training assay, wherein the training is based on the activity measurements stored in each training data set and an estimate of the amount of the target organism present in the training assays associated with each respective training data set, and Wherein the training assays include assays having different inhibition levels.

14. The system of claim 13, wherein the activity measurement is a time series measurement of light intensity collected over at least a portion of the nucleic acid amplification cycle.

15. The system of claim 13, wherein the activity measurements are time series measurements of light intensity collected within the nucleic acid amplification cycles.

16. A method of training a machine learning system to quantify the presence of a target organism in a biological assay, the method comprising: receiving a plurality of data sets, each data set associated with a biological assay, each data set comprising data collected by a detector during nucleic acid amplification of a target nucleic acid within the associated biological assay within one or more nucleic acid amplification cycles, wherein the data collected by the detector comprises activity measurements taken at different times during the one or more nucleic acid amplification cycles, wherein the target nucleic acid is associated with the target organism, and wherein the biological assays include biological assays with different inhibition levels; Each dataset includes: A first data subset, the first data subset includes measurement results obtained up to a first time point, the first time point being at time T max Before, the time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; A second data subset includes measurements taken after the first time point but before a second time point in the nucleic acid amplification cycle, the second time point occurring at T max After that; and a third data subset comprising measurements taken after the second time point in the nucleic acid amplification cycle; labeling each data set with an estimate of the amount of the target organism present in the associated biological assay; and A machine learning system is trained to estimate the amount of the target organism within a selected biological assay, the training being based on the activity measurements stored in each of the plurality of data sets and the estimate of the amount of the target organism present in the biological assay associated with each respective data set.

17. The method of claim 16, wherein the activity measurement is a time series measurement of light intensity collected over at least a portion of one or more of the nucleic acid amplification cycles.

18. The method of claim 16, wherein the activity measurement is a time series measurement of light intensity collected during one or more of the nucleic acid amplification cycles.

19. A non-transitory computer-readable medium storing instructions that, when executed by a processing circuit, cause the processing circuit to: receiving a plurality of data sets, each data set associated with a biological assay, each data set comprising data collected by a detector during nucleic acid amplification of a target nucleic acid within the associated biological assay within one or more nucleic acid amplification cycles, wherein the data comprises activity measurements taken at different times during the one or more nucleic acid amplification cycles, wherein the target nucleic acid is associated with a target organism, and wherein the biological assays include biological assays with different inhibition levels; Each dataset includes: A first data subset, the first data subset includes measurement results obtained up to a first time point, the first time point being at time T max Before, the time T max corresponding to the time in the nucleic acid amplification cycle when the measurement reaches a maximum amplitude; A second data subset includes measurements taken after the first time point but before a second time point in the nucleic acid amplification cycle, the second time point occurring at T max After that; and a third data subset comprising measurements taken after the second time point in the nucleic acid amplification cycle; and A machine learning system is trained to estimate the amount of the target organism within the selected biological assay, the training being based on the activity measurements stored in each data set and the estimate of the amount of the target organism present in the biological assay associated with each respective data set.

Images