System and method for pre-harvest detection of latent infection in plants

Abstract

To provide a system and method for pre-harvest detection of latent infection of plants.SOLUTION: Disclosed herein are systems and methods for identifying latent pre-harvest infection in plants. In one aspect, a method includes obtaining data describing expression levels of one or more biomarkers present in a plant, encoding the obtained data into a data structure for input into a machine learning model, and causing one or more computers to: Providing the encoded data structure as input to a machine learning model trained to generate output data indicative of a likelihood of the plant having latent infection based on processing of the encoded data structure, obtaining the generated output data indicative of the likelihood of the plant having latent infection, determining that the plant has latent infection based on the generated output data, and performing one or more operations to mitigate the infection of the plant.SELECTED DRAWING: None

Description

[Technical Field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is a continuation of U.S. Provisional Patent Application No. 63 / 11, filed November 5, 2020. This application claims priority to US Pat. No. 6,034,343, the disclosure of which is incorporated herein by reference in its entirety. is used as a reference.

[0002] Technical Field

[0002] This document relates to the pre-harvest prediction of latent infection in various types of plants (e.g., agricultural products). A device, system, and method for doing so are described. [Background technology]

[0003] background

[0003] Many plant products, such as fruits, can have high infection rates. Infection can occur when plant products become infected with pathogens during growth. For example, in avocado orchards, Approximately 20% to 40% of avocado trees, and therefore the avocados they produce, are yield-bearing. Pre-harvest plant products may be infected with anthracnose. Immature pre-harvest plant products may show symptoms of such infection. However, after the plant product has been harvested, ripened, and undergoes senescence, The produce may begin to show symptoms of infection, including stem rot, mildew, and streak / heart disease. There may be wilting and other characteristics that can lead to a detrimental loss of plant product quality.

[0004]

[0004] Harvesting involves removing the fruit from the parent tree, for example, through plucking or cutting. Picking, plucking and cutting can both expose the xylem elements of the fruit stalk to infection, and fungal infestation can occur. It may help the seeds to transfer into the xylem tissue of the fruit. The fungus grows on the fruit tree while the fruit is immature. They can colonize the surrounding tissues but remain dormant and actively grow as the fruit ripens. This can spread rapidly, causing cell damage and disease symptoms. ) are difficult post-harvest diseases: they originate on the farm, but still affect plant products (e.g. In addition, SER may not be detected until the product reaches the consumer. Infections such as these may have temporally and spatially independent severity. This means that early harvesting may be associated with the same This means that later harvests from the tree may have different SER incidence rates, and this incidence is not necessarily , may not be equal across neighboring trees. Summary of the Invention [Problem to be solved by the invention]

[0005]

[0005] Some pre-harvest and post-harvest mitigation methods involve the application of fungicidal chemicals to the crop before and / or after harvest. Repeated applications of fungicides can help to protect the parent tree from beneficial microorganisms. Repeated application of fungicides can have negative environmental consequences, including the deterioration of the flora. However, over time, this may lead to the development of resistant strains of pathogens. [Means for solving the problem]

[0006] overview

[0006] The present disclosure generally relates to the production of fruits, vegetables, seeds, and / or other products grown from plants that are evaluated. the potential for other plant products (e.g., leaves, stems, roots, etc.) derived from or otherwise derived from the plant being evaluated. It relates to the pre-harvest evaluation of plants (e.g., agricultural crops) to predict the occurrence of latent infection. In this study, machine learning models were used to predict plant potential using pre-harvest data about the plants and their predictive characteristics. The predictive features can be used to predict the likelihood of plant infection. These may include, but are not limited to, stage, age, hormones, dry matter content, environmental conditions, etc. do not have.

[0007]

[0007] In one exemplary implementation of the disclosed technique, fruit from the same trees throughout the seasons is harvested. By testing for rhizomes, it is possible to identify plants with a high incidence or incidence of one or more latent infections, such as stem rot (SER). Physiological and transcriptional information to profile fruit-producing trees under low A dataset of fruits including but not limited to avocados can be constructed. Latent infections such as may not become apparent until the fruit is well developed on the tree and ripens after harvest. However, fruit can have markers of infection as early as six months before peak harvest. Thus, the disclosed techniques may be used to control fungicide application, plant harvest schedules, and / or plant marketing. Applying pre-infestation corrections (e.g., mitigation treatments) to reduce the appearance of such infestation symptoms on plants. A plan to use predictive assays for preharvest detection of latent infection in plants with sufficient time to A platform can be provided.

[0008]

[0008] One or more embodiments described herein are directed to a method for identifying pre-harvest latent infection in plants. In one aspect, the method includes one or more computer-implemented methods. This describes the expression level of one or more infection biomarkers present in the above-ground parts of the plant. and inputting the data into a machine learning model by one or more computers. and encoding the acquired data into a data structure for processing the acquired data on one or more computers. The computer calculates the likelihood that a plant has a latent infection based on processing the encoded data structure. The encoded data is then fed to a machine learning model that is trained to generate output data shown in and providing as input a data structure for detecting the plant's latent state by one or more computers. and obtaining the generated output data indicative of the likelihood of having an infection; and determining that the plant has a latent infection based on the output data generated by the computer. and one or more computers to perform one or more processes to reduce latent infection in plants. and performing the procedure.

[0009] Another aspect is a method for detecting a virus, the method comprising: and corresponding systems, devices, and computers for performing the operations of the methods disclosed in this disclosure. These and other aspects optionally include one of the following features: For example, one or more computers may be used to detect latent infection in plants. Executing one or more actions to mitigate the above may be performed by one or more computers. Harvesting schedules for the above plants and / or one or more other plants of similar origin should be adjusted. determining that the harvesting schedule is adjusted by one or more computers; and storing data defining the rule in a memory device.

[0010] In some embodiments, one or more computers detect latent infection in plants. performing one or more actions to mitigate the When processed by the user device, the plant and one or more other plants of similar origin may be Send an alert to the user device informing the user that they should harvest. Generate an alert message that can be output to one or more computers. and transmitting the generated alert to a user device by the data processor. The one or more computers may then administer one or more treatments to reduce latent infection in the plant. The executing may include, by one or more computers, executing one or more including determining that the antimicrobial treatment can be administered to multiple other plants of similar origin. It is also possible.

[0011] In some embodiments, one or more other plants of similar origin are the same as the plant. In addition, one or more computers may be used to identify plants within the same zone. Implementing one or more procedures to mitigate latent infection of an object may involve one or more computers. When treated by the irrigation controller, the liquid containing the antimicrobial treatment is injected into the irrigation controller. generating one or more instructions that can be distributed to a user; A computer transmits one or more commands to an irrigation computer using one or more networks. As yet another example, the method may include transmitting the information to one or more controllers. Executing, by a computer, one or more actions to mitigate latent infection in the product may include: When processed by the robotic device by one or more computers, (i) navigating to a location associated with the plant; and (ii) connecting the plant and one or having a robotic device spray a liquid containing an antimicrobial treatment on several other locally grown plants. Antibacterial treatment can be initiated by a robotic device. generating one or more instructions; and executing, by one or more computers, one or more and transmitting one or more commands to the robotic device using the network. It can be done.

[0012] In some embodiments, one or more computers detect latent infection of the product. performing one or more actions to mitigate the When processed by the robotic device, the robotic device (i) detects a location associated with the plant. and (ii) from said plant and one or more other plants of similar origin. A harvesting operation that allows the robot device to harvest the plant product is generating one or more instructions capable of causing the execution of one of the following: or by multiple computers, using one or more networks, and transmitting the command to the robotic device.

[0013] In some embodiments, the data obtained includes (i) bark extracted from the plant. or (ii) a nucleic acid sequence of a sample of plant products extracted from the plant. Based on the output data generated by the nucleic acid sequencer Sometimes the data obtained is generated from (i) bark samples extracted from plants; or (ii) a nucleic acid sequencer of a sample of plant products extracted from the plant. Based on the output data generated by the nucleic acid sequencer based on sequencing by Additionally, data describing the expression level of one or more infection biomarkers in the plant can be generated. The data may contain a list of one or more variants. The data can describe the differences between the plant read sequences and the reference genome of a healthy plant. In some embodiments, the machine learning model is a binary logistic regression model, a logistic regression model, or a logistic regression model. Stick model trees, random forest classifiers, L2 regularization, partial least squares, or It may include one or more of one or more neural networks. , the plant products do not contain any visible signs of infection.

[0014]

[0014] In addition to the embodiments of the appended claims and the above embodiments, the following numbered The embodiment is also innovative.

[0015]

[0015] Embodiment 1 is a method for identifying pre-harvest latent infection in a plant, comprising: and generating data describing the expression levels of one or more infection biomarkers present in the plant. and obtaining an infection biomarker indicative of the likelihood of infection of the plant, the data being a list of the plant. and to obtain a genome sequence showing the differences between the plant and a reference genome of a healthy plant of the same species. and selecting, by the processor, one or more machine learning models based on the acquired data. and one or more machine learning models are used to compare other data with one or more other The data correlating with one or more identified infection biomarkers of the plant is used to identify the plant. Pre-trained to generate an output indicating the likelihood that an object will develop a latent infection before harvest One or more machine learning models are trained using the process, The process involves training a training dataset to one or more machine learning models: (i) one or (ii) non-invasive measurements of one or more other plants; or (iii) invasive measurements of one or more other plants. (iii) known plant information of one or more other plants; and (iv) one or more other inputting positive infection identification information of the plant; (i) to (iv) by one or more mechanistic Based on inputs into the learning model, one or more other plants may develop latent infection before harvest. and identifying a predictive likelihood for the model and selecting one or more machine learning models for runtime use. and selecting, by a processor, one or more machine learning models, including outputting the Based on applying the rule to the data, a result indicating the likelihood that a plant will develop a latent infection before harvest is obtained. generating a force and, based on the output exceeding a predetermined threshold range, determining that the plant has a latent infection; and reducing the latent infection in the plant by the processor. determining one or more treatments to be performed on the plant; and determining by the processor whether the plant has a latent infection. and outputting an indication of the occurrence of the condition and the one or more determined actions.

[0016]

[0016] Embodiment 2 is a method for controlling, by a processor, one or more compounds that reduce latent infection in plants. Identifying the treatment involves determining, for the plant and one or more other plants of similar origin, ) a change in harvest schedule to earlier in the growing season; (ii) a prescribed amount of insecticide before harvest; and (iii) instructions for applying the agent to the plant and one or more plants of similar origin to ensure that the plant is healthy. Identifying at least one of the times when each plant is nearing the end of its productive life, One or more machine learning models use data from the training dataset to The method of embodiment 1, which is pre-trained to identify i) to (iii). It is the law.

[0017]

[0017] Embodiment 3 is a method for growing one or more other plants of similar origin in the same zone as the plant. 3. The method of embodiment 1 or 2, wherein the plant is

[0018]

[0018] Embodiment 4 is a method for detecting, by a processor, an indication that a plant has a latent infection and outputting one or more determined actions when processed by the user device. and instructing the user device to perform one or more of the determined actions. Generate an alert message to notify the user, and and transmitting the alert message to the user device. This is one of the methods.

[0019]

[0019] Embodiment 5 is directed to a method for controlling, by a processor, one or more compounds that reduce latent infection in plants. The treatment decision may be based on the output produced and one or more other similar events prior to harvest. Any of embodiments 1-4, including determining that the source plant should be prescribed an antimicrobial treatment. One method is as follows.

[0020]

[0020] Embodiment 6 is directed to a method for controlling, by a processor, one or more compounds that reduce latent infection in plants. When determining the treatment is handled by an irrigation controller, the irrigation controller: generating instructions to automatically apply a liquid containing an antimicrobial treatment; and transmitting the instructions to an irrigation control. and transmitting the information to the server.

[0021]

[0021] Embodiment 7 is directed to a method for controlling, by a processor, one or more compounds that reduce latent infection in plants. When determining the treatment, the robotic device may include: (i) navigating to a location of a plant; (ii) administering an antimicrobial treatment to the plant and one or more other plants; generating instructions to spray plants of various origins and transmitting the instructions to a robotic device; and (iii) a method according to any one of the first to sixth embodiments, comprising:

[0022]

[0022] Embodiment 8 is directed to a method for producing, by a processor, one or more compounds that reduce latent infection in a plant. When determining the treatment, the robotic device may include: (i) navigate to the plant location; (ii) harvest the plant and the and one or more other plants of like origin to produce one or more plant products. and transmitting the command to the robotic device. This is one of the methods.

[0023]

[0023] In a ninth embodiment, the data obtained is a sequence of a plant product extracted from a plant. The nucleic acid sequence according to any one of embodiments 1 to 8, wherein the nucleic acid sequence is generated by a nucleic acid sequencer based on the sequence. This is the method described above.

[0024]

[0024] Embodiment 10 is a method for preparing a plant product comprising at least one sample of bark, leaves, flowers, and fruits. The method according to any one of embodiments 1 to 9, including one of the above.

[0025]

[0025] Embodiment 11 is an embodiment in which the plant product is sampled non-destructively from the plant. The method according to any one of aspects 1 to 10.

[0026]

[0026] Embodiment 12 is an embodiment in which the plant products include volatile components emitted from the plant. The method according to any one of modes 1 to 11.

[0027]

[0027] Embodiment 13 is a method for determining whether one or more machine learning models are binary logistic models. Regression model, logistic model tree, random forest classifier, L2 regularization, least squares, and convolutional neural networks (CNNs), The method according to any one of embodiments 1 to 12.

[0028]

[0028] Embodiment 14 is a method for growing plants according to embodiments 1 to 1, wherein the plants do not contain visible signs of infection. 3. The method according to any one of the preceding claims.

[0029]

[0029] Embodiment 15 is a method for processing, by a processor, the data to be input to one or more machine learning models. encoding the acquired data into a data structure; and and providing the generated data structure as an input to one or more machine learning models. 15. The method according to any one of embodiments 1 to 14, comprising:

[0030]

[0030] Embodiment 16 is a method for implementing one or more of the determined actions by a processor. 16. The method of any one of embodiments 1 to 15, further comprising reducing latent infection in plants by It is a method.

[0031]

[0031] In a seventeenth embodiment, the processor selects one or more machine learning models. and determining whether the one or more predictive features are further based on one or more predictive features identified from the acquired data. Alternatively, the plurality of predictive features may include a plant growing region, an environmental condition, a plant type, a plant growth stage, a plant Non-invasive measurements of plants, invasive measurements of plants, dry matter content, gene expression, plant volatile emissions 16. The method according to claim 1, further comprising at least one of: determining a growth zone; and determining a plant read sequence data. The method according to any one of the above.

[0032]

[0032] Embodiment 18 is a method for cultivating a plant in which known plant information is based on the growth patterns of one or more other plants. Historical information, length of growing season, soil conditions, precipitation level, amount of sunlight, environmental temperature, growing area, and plant 18. The method according to any one of embodiments 1 to 17, comprising at least one of the following types of products: .

[0033]

[0033] Embodiment 19 is a system for predicting latent infection in a plant, comprising: one or more and when executed by one or more processors, One or more devices storing instructions for causing the device to perform the method according to any one of the first to eighteenth embodiments. and a plurality of computer-readable storage devices.

[0034] The devices, systems, and techniques described herein have one or more of the following advantages: For example, early detection of infected plant products can provide a method for detecting the presence of infected plant products. Treatment and separation from uninfected plant product is possible, which reduces the spread of infection. , which can reduce waste of plant products and improve customer acceptance of the plant products. It can also reduce wastage of plant products and improve customer acceptance, thereby and other relevant users in the supply chain can provide higher quality plant products. , thereby enabling providers to differentiate themselves from other providers in the market. The technique can be applied early in plant growth, avoiding the onset of infection and wastage of plant produce. It can further provide for determining effective mitigation efforts to prevent the early onset of latent infection. The establishment of a timeline provides relevant stakeholders, such as farmers, with ample time to determine and apply mitigation efforts. Additionally, sampling plants before harvesting can provide a convenient time frame. Determine and apply mitigation efforts to eliminate the potential loss of plant products during the plant maturation process. or can be reduced, so that plants (e.g. trees) are available for sampling. Loss of plant produce is not a burden to farmers.

[0035]

[0035] As another example, the disclosed techniques may collect various data and / or provides a combination of features that allow accurate and timely pre-harvest prediction of latent infection. Humans may use invasive measurements to predict the onset of latent infection. Various types of data collection on plants during the plant growth period, such as collection of plant values ​​and non-invasive measurements are neither effective nor accurate in collecting and assessing disclosed data. The techniques described are not intended to be used by humans without the present invention to accurately predict latent infection in plants before harvest. collection of various types of data that may be difficult, error-prone, and / or impossible to obtain; Provides collection, synthesis, and analysis.

[0036]

[0036] Furthermore, the present disclosure provides a method for detecting latent infection in a zone of plants of similar origin. In some embodiments, the present disclosure provides a method for producing a plant product that requires less disruption. Rather than sampling the plant product itself, only the above-ground parts of the plant are sampled. By taking the plant product, the destruction of the plant product can be reduced. In that case, the operation of the present disclosure is , may be performed using samples of the above-ground parts of the plant. The present disclosure provides a method for detecting latent infection in plants and reducing and potentially eliminating latent infection before it becomes apparent. and providing the benefit of preserving plant produce by initiating one or more remedial actions of This can improve the availability of plant products and customer satisfaction. Such remediation actions can include, for example, harvesting plant products from plants and refining them with other plants of similar origin. a user or robot to harvest plant produce from one or more other plants within the plant zone. In other embodiments, the harvest schedule may include instructing the device to: This may be updated to achieve similar benefits, which may provide one or more economic advantages of the present disclosure. can be provided to users.

[0037] The details of one or more implementations are set forth in the accompanying drawings and the description below. Other features and advantages will be apparent from the description and drawings, and from the claims. [Brief explanation of the drawings]

[0038] BRIEF DESCRIPTION OF THE DRAWINGS [Figure 1A]

[0038] A conceptual diagram of an example system for detecting latent infection in plants. [Figure 1B]

[0039] 1 is a flow chart of a process for detecting latent infection in plants. [Figure 2]

[0040] A conceptual diagram for predicting the likelihood of infection in plants. [Figure 3]

[0041] FIG. 1 is a conceptual diagram of the training of one or more models to predict infection likelihood in plants. [Figure 4]

[0042] 1 is a flowchart of a process for predicting the likelihood of infection in a plant. [Figure 5]

[0043] FIG. 1 is a system diagram of one or more components that can be used to implement the techniques described herein. [Figure 6A]

[0044] 1 illustrates an example process for collecting plant data for use in conjunction with the disclosed techniques. [Figure 6B]

[0044] An example process for collecting plant data is shown for use in conjunction with the disclosed techniques. [Figure 6C]

[0044] An example process for collecting plant data is shown for use in conjunction with the disclosed techniques. [Figure 7A]

[0045] 7 illustrates a graphical representation of example physical attributes of the plant of FIG. 6 that may be collected at different times. [Figure 7B] FIG. 7 shows a graphical representation of example physical attributes of the plant of FIG. 6 that may be collected at different times. [Figure 8A]

[0046] A biplot of principal component analysis of normalized gene expression across all time periods of data collection for the plants in Figures 6A-6C is shown. [Figure 8B]

[0046] Figures 6A-6C show biplots of principal component analysis of normalized gene expression across all time periods of data collection for the plants. [Figure 8C]

[0046] Figures 6A-6C show biplots of principal component analysis of normalized gene expression across all time periods of data collection for the plants. [Figure 9A]

[0047] Volcano plots of normalized gene expression for the plants in Figures 6A-6C are shown. [Figure 9B]

[0047] Volcano plots of normalized gene expression for the plants in Figures 6A-6C are shown. [Figure 9C]

[0047] Volcano plots of normalized gene expression for the plants in Figures 6A-6C are shown. [Figure 10A]

[0048] Clustered heat maps of normalized gene expression for genes analyzed at each collection time for plants in Figures 6A-6C are shown. [Figure 10B]

[0048] Figures 6A-6C show clustered heat maps of normalized gene expression for genes analyzed at each collection time for plants. [Figure 10C]

[0048] Figures 6A-6C show clustered heat maps of normalized gene expression for genes analyzed at each collection time for plants. [Figure 10D]

[0048] Figures 6A-6C show clustered heat maps of normalized gene expression for genes analyzed at each collection time for plants. [Figure 11]

[0049] FIG. 1 is a block diagram of system components that can be used to implement a system for detecting latent infection. DETAILED DESCRIPTION OF THE INVENTION

[0039]

[0050] Like reference symbols in the various drawings indicate like elements.

[0040] DETAILED DESCRIPTION OF ILLUSTRATIVE EMBODIMENTS

[0051] The present disclosure generally relates to systems, methods, and computers for detecting latent infection in plants. Many plants produce fruit periodically, often for decades or centuries. Live productively. Throughout this long lifespan, plants maintain populations of beneficial and pathogenic microorganisms. They can interact with and respond to the surrounding ecosystems they can contain. The detection of and response to pathogenic microorganisms by plants involves genes involved in the plant's innate immune system. The disclosed technique can be used to analyze the genetic information from trees. We profiled the transcriptomes of various plant products from different types of plants, including the fruits of By detecting pre-harvest symptoms of latent infections such as stem rot (SER), pre-harvest translocation Such latent infections can be detected in the fruit. may not become apparent until mature on the tree, harvested, and / or ripe. Fruits and other plant products provide markers of infection as early as six months before peak harvest season. This information can be used to generate detection assays that can be used to identify a variety of species. application of fungicides to reduce the manifestation of latent infection in plants, and harvest schedules for plant products. High incidence trees with ample time to correct the leaflet and / or plant product destination information. and prediction of short trees becomes possible.

[0041]

[0052] The present disclosure provides a method for generating plant traits based on machine learning model processing of input data representing plant traits. A trained method capable of generating output data indicative of the likelihood that an object has a latent infection. A machine learning model may also be provided. In some embodiments, the plant features are and data representing the expression level of one or more biomarkers detected in the sample. If the plant is suspected to have a latent infection, the disclosed techniques can detect the detected latent infection. determining one or more actions designed to mitigate the infection, and optionally Or it may be provided to perform multiple operations.

[0042]

[0053] Referring now to the drawings, FIG. 1A illustrates an example system 100 for detecting latent infection in plants. The system 100 includes a user device 110, one or more robot devices, and devices 120, 121, 122, 124, 125, 126, nucleic acid sequencer 130, a computer 140 (e.g., a computer system), a network 150, and one or more robotic device charging stations 160, 162, 164, an irrigation system 170, and System 100 may include these components: and the plant, plant product, or both. The functionality described herein can be achieved by detecting latent infection of In this case, a nucleic acid sequencer may not be necessary. Instead, a qP Further, in some embodiments, the sequencer 130 and the computer 1 40 can be part of the same system.

[0043]

[0054] Referring to FIG. 1A, the implementation of system 100 may, in some embodiments, include: The sample 105 is a sample 122 of plant product of plant 202-1 in Zone 2, and the assignment of plant 202-1 The sampling may involve extracting samples of any above-ground part of the plant, or both. This can be done before harvest, which is sometimes done before the plant products of the plants being sampled are collected. The disclosed techniques can be used to treat sausages that have been sown for six months or more before being harvested. Pulled plants are harvested while they are undergoing their maturation process and before they are harvested (e.g., fully grown). As a result, the disclosed techniques can be used to prevent latent infections. sampled transplants early enough that efforts can be made to mitigate potential outbreaks. This can be performed to predict the likelihood of infection of plants and other plants of similar origin before harvest.

[0044]

[0055] Sample 122 contains genetic data of the plant products of the sampled plant. In some embodiments, the sample 122 may be a sample of fruit, leaves, or wood of the plant 202-1. The sample may be derived from the skin, plant tissue, or other plant product of plant 202-1. 122 may contain other information about the plant products of the sampled plant 202-1. For example, the sample 122 may be analyzed by hardness and / or penetrometer measurements, volatile metabolism, or the like. and other compounds that can be detected non-destructively and / or destructively. .

[0045]

[0056] In some embodiments, the sample 122 is collected by one or more drones 120 or can be extracted by other types of collection devices or mechanisms. Examples of Collection Devices The method may include a stationary probe device and / or an automatic stationary sampler. Collecting the sample 122 using one or more of the types of collection devices described herein. In some embodiments, the sample 122 contains a nucleic acid sequence to be sequenced. The sample 122 may be pre-treated to prepare it for input to the sensor 130. The processing may involve, for example, processing the sample 122 to obtain a nucleic acid sample from the sample 122. barcoding the sample, or any combination thereof. In some embodiments, sample 122 is the above-ground portion of a plant, such as plant 202-1. In such an embodiment, the above-ground portion of the plant may be cultured using the techniques of the present disclosure. One or more infections of nucleic acids derived from the sample 122 of the plant 202-1 are analyzed. The expression level of the biomarker can then be determined. Based on the treatment of only the aboveground parts of 02-1, the plant products (Fig. 1A) For example, the zones are shown using 102-1 to 102-z, which are shown in zones 1 to n. n is any positive integer, x is any positive integer, y is any positive integer, and z is any positive integer) has or may develop latent infection In some embodiments, the above-ground parts of the plant and the plant products can be predicted. A sample of can be used.

[0046]

[0057] The nucleic acid sequencer 130 receives nucleic acids derived from the sample 122 and The output data can be generated by processing the nucleic acid (e.g., The sequence data can be referred to as read sequence data 132, which represents the order of nucleotides in a sequence (e.g., DNA). In some embodiments, the read sequence data 132 includes guanine (G), cytosine (C), One or more nucleotide salts containing adenine (A) and thymine (T) in any combination The group can be used to include nucleotides within the nucleic acids of the sample 122. In an embodiment, the read sequence data 132 is a read sequence generated by the nucleic acid sequencer 130. Sequence data generated by transcribing into cDNA representing the RNA of sample 122 In such an embodiment, the read sequence data can be divided into G, C, A, and uracil ( U) in any combination.

[0047]

[0058] The read sequence data 132 is collected using one or more networks 150. The one or more networks 150 may be wired or Ethernet network, optical network, wireless network, WAN, LAN, A cellular network, a Wi-Fi network, the Internet, or any combination thereof In some embodiments, the nucleic acid sequencer 130 may include an Ethernet a t cable, a USB-C cable, or a cable between the nucleic acid sequencer 130 and the computer 140 Application servers using any other form of direct connection that facilitates data communication between The computer 140 receives the read sequence data 132 as input. The biomarker expression engine 151 can provide the biomarker expression information to the biomarker expression engine 151. 51 may be part of a computer 140, as further described with reference to FIG. Cut.

[0048]

[0059] In some embodiments, the data output by the nucleic acid sequencer 130 is Epigenetic data can be derived from nucleic acid sequences. The nucleic acid sequencer 130 performs sequencing of the sample 122 based on the sequencing of the sample 122 by the nucleic acid sequencer 130. The resulting read sequence data 132 may include base call (or nucleotide) changes. Changes to the nucleic acids in the sample (e.g., chemical modifications, changes to the structure of DNA, etc.) that cannot be detected In some embodiments, such epigenetic data can include: For example, it may be generated by a long-read sequencer. If no network data is used, this includes but is not limited to short-read sequencers. Other nucleic acid sequencers can be used that do not require epigenetic sequencing. The epigenetic sequences described herein are merely examples of sequences that can be used to detect Any other sequencer may be used to generate the data.

[0049]

[0060] The biomarker expression engine 151 receives the read sequence data 132 and Publication No. 2021 / 092251A1 PREDICTION OF INFECTION IN PLANT PRODUCT As described in S, the expression level of one or more infection biomarkers, housekeeping The system can be configured to determine the level of a biomarker, or both, which can be used to determine overall The biomarker expression engine 151 may include one or more is the expression level of multiple infection biomarkers, the level of housekeeping biomarkers, or generating output data 151a based on processing of read sequence data 132 representing both In some embodiments, the expression level of each biomarker can be determined by a read sequence device. The number of occurrences of each gene in 132 may be included. The number of occurrences may be expressed as copy number.

[0050]

[0061] In this example, the nucleic acid sequencing device 130 sequences the sample 122. The biomarkers are then sampled and used to generate read sequence data 132. The expression engine 151 calculates the expression levels of one or more infection biomarkers, house key In such an embodiment, the level of a ping biomarker, or both, can be determined. This is the gene expression represented as copy number by the biomarker expression engine 151. The disclosure is not so limited, and some embodiments may include determining the number of occurrences of In this case, the nucleic acid sequencer 130 may not be necessary.

[0051]

[0062] For example, in some embodiments, a quantitative polymerase chain reaction device (qPC Using R) machines and microarrays to identify gene occurrences in sample 122 In such an embodiment, the primers can be configured in a qPCR machine to measure the amount of the sample 12 2. qPCR machines can identify specific gene sequences within a target gene. The qPCR method can be configured to amplify specific gene sequences identified by the method. The output is a count of only the specific subset of genes identified by the assay. qPCR machines can be used to sequence complete genomes using one or more types of sequencing devices. or transcriptome sequencing. The counts output by the computer 140, the biomarker expression engine 151, and in some embodiments, may be entered without first being provided to the biomarker expression engine 151. The force generation engine 152 may be provided to the user device 110, the computer 140 or configured to communicate with the computer 140 using a network 150. Implemented in any other computer, device, and / or system capable of It is possible.

[0052]

[0063] The input generation engine 152 can be part of the computer 140. Expression levels of one or more infection biomarkers, levels of housekeeping biomarkers The input generation engine 15 can receive output data 151a representing the input, output, or both. 2 is used to input data into a latent infection prediction engine 153 based on the received data 151a. In some embodiments, this can generate an input data structure for the infectious bio Data representing the expression levels of markers, housekeeping biomarkers, or both are collected on the array. For example, the encoded array data structure may include: levels of one or more infection biomarkers, housekeeping biomarkers, or both The input generation engine 152 may include a binary string representing the encoded input data structure. The structure 152a can be provided as an input to a latent infection prediction engine 153.

[0053]

[0064] The latent infection prediction engine 153 can be part of the computer 140. , and can receive input data structure 152a. PREDICTION OF INFECTION IN PLANT PRODUCT, International Publication No. 2021 / 092251A1 One or more trained using the training process described in the CTS or may otherwise include one or more of such machine learning models. The present invention relates to a method for manufacturing a computer system, a method for manufacturing a computer-implemented computer system, and a computer-implemented system. The training model is based on one or more infection biomarkers, housekeeping biomarkers, or based on processing of an input data structure 152a that encodes data representing both expression levels. The one or more machine learning models can be sampled and used to generate output data. Destructive (e.g., invasive) measurements taken from plants 202-1, sampled Non-destructive (e.g., non-invasive) measurements taken from the plants 202-1, environmental conditions, data (e.g., weather patterns, rainfall levels, number of sunny days, other growing conditions), and / or Includes data on the geographic area where the sampled plants 202-1 are grown and harvested. Additional input data may also be provided, including but not limited to: and / or using the input data structure 152a, one or more machine learning models may Pulled plants 202-1 and Zone 2 or other zones within the growing area (e.g., field) It is possible to predict the likelihood of developing latent infection in other plants of similar origin within the same population. It can be in the form of force data.

[0054]

[0065] Therefore, the output data is one plant 202-1 from which sample 122 was derived. or have developed, are developing, and / or may be developing multiple latent infections The latent infection prediction engine 153 can represent the likelihood 153a. / 092251A1, PREDICTION OF INFECTION IN PLANT PRODUCTS The functions described by the model can be performed, which are incorporated herein by reference in their entirety. The output data 153a generated by the latent infection prediction engine 153 is In some embodiments, the latent infection detection module 154 may provide the latent infection detection The dye prediction engine 153 uses binary logistic regression models, logistic model tools, L2 regularization, or one or more neural networks. The model may include one or more of a linear network, a partial least squares model, and / or a partial least squares model.

[0055]

[0066] The latent infection detection module 154 can be part of the computer 140. and evaluating the output data 153a to determine whether mitigation efforts may be necessary. By way of example, the output data 153a may be measured against one or more predetermined thresholds. It is determined that the plant 202-1 from which the sample 122 was derived developed one or more latent infections. Whether the output data 153a indicates that a disease is occurring, developing, and / or may be developing. In some embodiments, the latent infection detection module 154 can determine whether: International Publication No. 2021 / 092251A1, PREDICTION OF INFECTION IN PLANT PRO Output data 153a, which is the same as or similar to the output data of the prediction module described in DUCTS can be evaluated, which is incorporated herein by reference in its entirety. Output Data If the latent infection detection module 154 determines that 153a does not indicate the presence of a latent infection, For example, the techniques described herein may be transmitted at 156. If the latent infection detection module 154 determines that the output data 153a indicates the presence of a latent infection, If so, the latent infection detection module 154 determines, suggests, and takes one or more mitigating actions. and / or may provide instructions 154a to the mitigation engine 155 for execution.

[0056]

[0067] The mitigation engine 155 can be part of the computer 140 and can include one or more may initiate the determination, suggestion, and / or execution (e.g., enforcement) of multiple mitigating actions. These operations can be performed in a variety of ways. For example, in some implementations, The detection engine 155 receives instructions 154a, output data 153a, and one or more infected viruses. Expression levels of biomarkers, expression levels of one or more housekeeping biomarkers , or a combination thereof, based on the sampled plants 202-1 and / or one or determine that the harvest schedule for multiple other plants of similar origin can be adjusted In such an embodiment, the abatement engine 155 may 02-1, plant products of the plant, plants and / or plant products of similar origin, or combinations thereof The database storing the harvest schedule of the farmer is accessed, and the harvest schedule is calculated based on the output data 153a. In such an embodiment, the harvesting schedule can be adjusted accordingly. The output 155a of 155 is the data written to the database to adjust the harvest schedule. In some embodiments, adjustments to the harvest schedule may include sample Post-harvest export of harvested plant products from harvested plants 202-1 and / or plants of similar origin This may include adjusting the route of transmission. For example, if the predicted likelihood of latent infection is high (e.g. (e.g., the predicted likelihood of latent infection exceeds some predetermined threshold or range) Plant products from 202-1 and / or other plants of similar origin stay fresh longer. , can survive further circulation, have a low predicted likelihood of latent infection, or are Compared to other plants, the plant products are delivered to geographic areas closer to the place of harvest. In some embodiments, adjustments to the harvest schedule can be made easily. The decision engine 155 may determine whether the user device 110 or another relevant stakeholder The relevant stakeholders can then make the adjustments. The adjustments can be made based on the relevant may be implemented by the stakeholder and / or by the mitigation engine 155 or the entire disclosure. may be performed semi-automatically by one or more other components described throughout .

[0057]

[0068] For example, in some embodiments, the abatement engine 155 may be configured to: To avoid this, instructions 154a, output data 153a, and one or more infected bio the expression level of the marker, the expression level of one or more housekeeping biomarkers; or combinations thereof, based on plants, e.g., 202-1, plant products, plants of similar origin or The plant may determine that a zone of plant produce should be harvested immediately. In an embodiment, the abatement engine 155, when processed by the user device, plant product, and / or one or more other plants or plant products of similar origin should be harvested. issue an alert to the user device 110 informing the user 105 of the user device 110 that In such cases, the output The output 155a may include a generated alert. The computer 140 then , the generated alert can be sent to the user device 110.

[0058]

[0069] The mitigation engine 155 may communicate with the user 105, other relevant stakeholders, and / or or one that can be taken by another device / robot (e.g., drone 120) A number of other actions may also be determined. For example, as further described below, the mitigation engine 1 55 preserves such plants and / or products and prevents latent infection during the growing season before harvest. To produce one or more products, the plant 202-1, the plant product, and / or one or more It may be determined that the product can be sprayed on other plants or plant products of similar duration.

[0059]

[0070] Mitigation engine 155 is responsible for plant 202-1 and other plants of similar origin within the same zone. The plant may be harvested in several ways. For example, in some embodiments, the plant may be harvested using a reduced energy engine. The drone 155 deploys one or more robotic devices (e.g., drone 120) to Plant 202-1 and other plants from the same zone, such as Zone 2 in Figure 1. In such an embodiment, the abatement engine 155 When processed by the robotic device, the robotic device (i) detects a plant-associated event. and (ii) initiating a harvesting operation by the robotic device. The robotic device may generate one or more instructions to cause the robotic device to perform harvesting. The initial step is to obtain at least one genotype from the plant and one or more other plants of similar origin in zone 2. and initiating the robotic device to perform one or more actions associated with harvesting the plant product. In such an embodiment, the output 155 of the mitigation engine 155 a may include instructions that cause the robotic device to navigate and begin harvesting. In some implementations, the mitigation engine 155 may be configured to: Using the location data of drone 120 that collected the sample of plant 202-1, The location of the plant and one or more other plants of similar origin within zone 2 can be identified. The mitigation engine 155 uses the network 150 to communicate with the robotic devices 124, 12 5, and / or 126, and / or robotic device 120, or any other type of automated One or more commands can be sent to the harvesting device. Although shown as a robotic device, the disclosure is not so limited. The robot devices 124, 125, and / or 126 are robots that move on the ground as rovers. Any robotic device, including a robotic device, that navigates via aerial flight. The device may include a chair, or any combination thereof.

[0060]

[0071] As yet another example, in some embodiments, the mitigation engine 155 may 4a, output data 153a, expression levels of one or more infection biomarkers, or based on the expression levels of multiple housekeeping biomarkers, or a combination thereof. It may be determined that an antimicrobial treatment should be administered to one or more other plants of similar origin. In some embodiments, the antimicrobial treatment can be any of the antimicrobial treatments described further below (see, e.g., Table 1) and in WO 02 / 04444. No. 2021 / 092251A1, PREDICTION OF INFECTION IN PLANT PRODUCTS and the like, which are incorporated herein by reference in their entirety. The abatement engine 155 can administer the antimicrobial treatment in many different ways.

[0061]

[0072] For example, in some embodiments, the mitigation engine 155 may to the remote irrigation controller 171 using One or more devices that can activate the water supply system 170 and deploy antimicrobial treatment. The instructions can be generated by, for example, the irrigation controller 171 or the abatement engine 155. can command valve 172 to close and valve 174 to open. 71 then treats plant 202-1 and one or more plants in the same zone, such as zone 2, with an antimicrobial treatment. sprayed on multiple local plants to reduce detected or otherwise predicted latent infections; In another embodiment, the mitigation engine 155 may include instructions 154a, output data 153, and a. Expression levels of one or more infection biomarkers, one or more housekeeping Plant 202-1 and Zoo 202-2 were identified based on the expression levels of the markers, or a combination thereof. Other plants of similar origin in Zone 2 only need to be watered and no antibacterial treatment is required. In such an embodiment, the mitigation engine 155 may determine that the irrigation controller 1 71 (e.g., valves 174, 172) may be connected to a water tank, public water source, well water source, or a combination thereof. 178. can.

[0062]

[0073] The abatement engine 155 may also distribute the antimicrobial solution in other ways. For example, in some embodiments, the abatement engine 155 may be configured to initiate one or more antimicrobial actions. Multiple robotic devices can be commanded. In such an embodiment, Each robotic device, such as the antimicrobial solution dispenser 120, is connected to a tank of antimicrobial solution and a spraying device 12 In such an embodiment, the robotic device 120 may include a mitigation Receives the commands generated and transmitted by the engine 155, processes the commands, and controls the plants and the same zone. Move to areas associated with other plants of similar origin in the garden and use sprays 121a and 122a. The antimicrobial solution can then be administered to the plant and other plants of similar origin in the same zone.

[0063]

[0074] The above examples involve analyzing read sequences to identify gene counts and / or qPC One or more infection biomarkers, identified using the R machine as gene counts, Alternatively, multiple housekeeping biomarkers, or both, may be detected. Instead, the system 100 may be used to generate a prediction engine 153 Other types of biomarkers can be identified and counted for input into the In some embodiments, system 100 is used to detect genes that may be indicative of drought stress. can be detected and counted as indicative of drought stress. Examples of the types of genes that can be used are described further below in Table 1. The expression levels of drought stress were associated with the expression of infection biomarkers, housekeeping biomarkers, or both. The expression levels of the two genes can be analyzed using the system 100 in the same or similar manner. Once the system 100 detects the threshold expression level of drought stress, the system 100 to use the mitigation engine 155 to reduce the damage to a particular plant and / or one or more plants in the same zone. The irrigation system 171 can be instructed to perform watering of plants of similar origin. .

[0064]

[0075] FIG. 1B is a flow chart of a process 180 for detecting latent infection in plants. Process 180 is illustrated and described throughout this disclosure (see, e.g., FIG. 1A, FIG. 5). ) can be executed by computer system 140. Process 180 and / or One or more blocks of process 180 may be implemented in other computing systems, devices, or computers. Networks of computers / devices, cloud-based systems, and / or cloud-based For purposes of illustration, process 180 may be performed by a computer. It is explained from a data system perspective.

[0065]

[0076] Referring to process 180 of FIG. 1B, a computer system and receiving data describing the expression levels of infection biomarkers in plants within the system. (182) With reference to Figure 1A and as explained throughout this specification, the data , one or more plants collected by a user (e.g., a human worker) in the field / farm The sample can be collected by a device such as a drone or other automated collection device. They can also be collected by vise.

[0066]

[0077] The computer system encodes the data for input into one or more models. The data can be input into one or more machine learning models. The biomarker expression engine 151 and the input generation engine 152 see legend in FIG. 1A.

[0067]

[0078] The computer system was designed to predict the likelihood of latent infection in plants at 186. At least one model trained as described further below can be selected. As such, different species may be present based on the type of plant, the stage of plant development, and / or one or more environmental conditions. Different models can be selected for application. The models are applied sequentially. Models can also be applied in parallel.

[0068]

[0079] In 188, the computer system is coded to the selected model. The computer system then executes the modeling at 190. An output can be received from the test module, the output indicating that the plant has or is suspected to have a latent infection. The value can be a number (e.g., 1 to 10). On a scale of 0, 1 being the least likely to develop a latent infection and 100 being the most likely to develop a latent infection. (Most likely a string value and / or a Boolean value.)

[0069]

[0080] In 192, the computer system calculates a predicted likelihood of latent infection in a plant (e.g., It is possible to determine whether the predicted likelihood (output from the model) exceeds a predetermined threshold range. If the value exceeds a predetermined threshold range, the computer system detects the plant and other plants in the particular zone. One or more mitigating actions can be determined that mitigate latent infection in plants of similar origin. (194). In other words, the computer system measures plant waste and / or plant or act early during the growth of the plant to avoid the occurrence of losses of its plant products. It can be determined that the plant has a sufficiently high likelihood of developing latent infection to prevent the disease.

[0070]

[0081] In some embodiments, the computer system performs one or more mitigation actions. The user can then select which The user can choose to take mitigation action. The mitigation action can be Mitigation actions can be performed automatically by a computer system. Mitigation actions may be taken when the user provides input to the computer indicating the selection of such mitigation actions. It can also be executed semi-automatically by a computer system when provided to the computer system. do.

[0071]

[0082] If the predicted likelihood of latent infection of the plant does not exceed the predetermined threshold range, process 180 In other words, the computer system can determine whether a particular plant is At this point in the growing season, the incidence of latent infection is high enough to trigger some kind of mitigation action. It can be determined that there is no likelihood of disease.

[0072]

[0083] FIG. 2 is a conceptual diagram of predicting the likelihood of infection in plants. 0, the user device 110, the plant data store 230, and the model data store 240 , and can communicate (e.g., wired or wireless) via a network 150. The harvester 105, in step A, extracts pre-harvest data from plants in a field / farm zone. Pre-harvest collection can occur as early as six months before harvest. Pre-harvest collection may occur at one or more time intervals during the plant's growing season, for example Pre-harvest collection and subsequent infection likelihood prediction can be performed every two months until harvest. Therefore, pre-harvest collection and infection likelihood prediction can be performed up to six months before harvest. At this point, mitigation actions can be determined and applied to the plant and / or plants of similar origin. At this point, additional pre-harvest collection and / or forecasting is required for the remainder of the growing season prior to harvest. You can also determine whether or not you can get it.

[0073]

[0084] Pre-harvest collection may include any of the techniques described herein (e.g., see FIG. For example, the sampler 105 (e.g., a user) walks around the plant and In some embodiments, a subset of plant products can be collected from the sample. The sampler 105 selects four plants from the plants being sampled and tested for likelihood of latent infection. The sampler 105 is ideal for predicting the likelihood of latent infection. Any other quantity of plant product may be harvested from the plant. Dependent on the type of plant, the particular area in which the plant grows, and / or one or more other environmental conditions Therefore, the collection of plant products can be limited to certain zones of the field / plot. This can be used to determine whether a particular plant (e.g., a tree) is at risk of infection. Cut.

[0074]

[0085] As another example, collection may involve detecting plant volatile metabolites using non-destructive techniques. Such data can include information about how plants respond to environmental conditions. This can also provide insight into whether the plant has a latent infection and and / or may be another indicator of whether or not a person is likely to develop a latent infection. The harvester 105 may be configured to place a volatile collection device (e.g., NIR device, electronic nose) in the vicinity of the plant product. or other equipment) to collect the emitted volatile substances. Including, but not limited to, electronic noses and / or NIR devices on automated robotic devices Other field-deployable instruments may be used to collect and measure volatile compounds emitted from plant products. You can also do this.

[0075]

[0086] Any combination of pre-harvest collection can be performed. It can also depend on which model is used to predict the likelihood of latent infection, which is , may further depend on the plant species, growing region, and / or environmental conditions. In an embodiment, volatiles can be collected from the plant products of the plant and can be collected in a quantity from the plant. The plant products may also be harvested and transported to a laboratory or facility for further analysis.

[0076]

[0087] Once the pre-harvest data is collected (Step A), the data is stored in a computer system. 140 (Step B). As described herein, the data may be sent to the It can include sequencing data, which can be obtained by a nucleic acid sequencer or other sequencing device. The data can be transmitted to a device (see, for example, Figure 1A). In some embodiments, data may be collected and / or harvested from plants. The present invention may include other measurements, non-destructively obtained and / or destructively obtained, of the plant product.

[0077]

[0088] The computer system 140 receives plant information from the plant data store 230. Information can be collected about plants being tested for infection (Step C). The information may be specific to the type of plant and / or plant product that was collected. The information can be region specific. The information can be based on the current environmental conditions and / or the plant's growth prior to harvest. Step C may include one or more of the environmental conditions described in Figure 2. can be performed before, between, and / or after multiple steps at any time. Step C can be executed simultaneously with the execution of step D.

[0078]

[0089] The computer system 140 may process the data in step D. The processing of the data can be performed as described in FIG. 1A. ,By encoding, the collected harvest data is prepared and ,used as input to one or more models. The processing of the data may include providing the data for input to one or more models. This may also include preparing plant information based on the results of the survey.

[0079]

[0090] The computer system extracts data from the model data store 240 based on the plant information. The computer system may also receive a model (Step E). It is possible to determine how many models to apply to predict the degree of The system selects which model to use based on plant information such as plant type and / or environmental conditions. You can also decide whether to

[0080]

[0091] The computer system applies the selected model in step F. For example, the computer system may process collected pre-harvest data and / or plant data. Object information can be provided as input to the model (eg, in an encoding structure).

[0081]

[0092] Therefore, in step G, the computer system calculates the likelihood of infection of the plant. For example, the model can provide an output in the form of a value indicating the likelihood of infection of a plant. The computer system can then determine whether the value falls within some predetermined threshold range. It can be determined whether the threshold range is exceeded. Exceeding the threshold range means that the plant is infected. If the value does not exceed the threshold range, the plant is likely to develop infection. The likelihood of developing the disease is low.

[0082]

[0093] Based on the prediction in step G, the computer system 140 optionally Finally, mitigation actions can be determined in step H. For example, decisions, suggestions, and and / or mitigating actions that can be implemented (e.g., manually, automatically, and / or semi-automatically). See Table 1 below for further discussion of types.

[0083]

[0094] The computer system 140 may be configured to predict the likelihood of infection and / or determine mitigation actions. The user device 110 can then transmit the operation to the user device 110 (Step I). 0 can output predictions and / or mitigation actions to be presented to relevant users (e.g., Step J). For example, the user device 110 may be a mobile phone with a display screen, a computer The predictive and / or mitigating actions may include: It can be presented on a display screen in a graphical user interface (GUI). Such information can be viewed and an action can be taken in response to viewing the information. For example, The associated user selects one or more mitigation actions to be performed from among the presented mitigation actions. It is possible.

[0084]

[0095] The user device 110 and / or computer system 140 then optionally Optionally, perform mitigating actions (step K). In some implementations, the relevant user: At the user device 110, the mitigation action to be taken can be selected. The notification may be sent to a computer system 140. The computer system then and instructing one or more robotic devices to perform notification and user-selected mitigation actions. In some implementations, the user device 110 may transmit a mitigation action to If to be executed, the instructions can be sent to one or more robotic devices. At times, the instructions may be transmitted to one or more other user devices, and such devices Inform the user of the computer that one or more mitigating actions should be taken. In some embodiments, the user device 110 and / or the computer system 140 may In step K, the selection is not received and / or the associated user is selected at the user device 110. Mitigating actions may be taken without user approval.

[0085] [Table 1]

[0086] [Table 2]

[0087] [Table 3]

[0088] [Table 4]

[0089] [Table 5]

[0090]

[0096] Examples of gene categories in Table 1 are PAL: phenylpropanoid ammonia CHS: chalcone synthase; PER: peroxidase; HSP: heat shock protein Protein; WRKY: WRKY transcription factor; NAC: NAC transcription factor; ERF: ethylene response factor response factor; ETR: ethylene transcription factor; EIN: ethylene-insensitive protein; ABA: ethylene bucidic acid; NCED: 9-cis-epoxycarotenoid dioxygenase; ZEP: Zeaxanthin epoxidase; ABI: Abscisic acid-insensitive transcription factor; JMT: Ja salicylate methyltransferase;SAMT: salicylate methyltransferase; and PIN: PIN-FORMED proteins. One or more other gene categories may be identified. It can also be used in conjunction with the techniques shown.

[0091]

[0097] FIG. 3 shows the steps of training one or more models to predict the likelihood of infection in plants. The model is a conceptual diagram, which may be used in conjunction with, for example, any of the techniques described in this document and in WO 2004 / 024444. No. 021 / 092251A1, PREDICTION OF INFECTION IN PLANT PRODUCTS You can train using any of a variety of techniques, including any of the techniques used and the like, which is incorporated herein by reference in its entirety and is the property of one or more of the same inventors. It has been filed by several people.

[0092]

[0098] The training of one or more models is performed by a computer system 140. Although shown as being performed, the training may be performed on one or more other computers. systems, devices, networks, cloud-based systems, and / or clouds It can also be performed by a service based on the

[0093]

[0099] Referring to FIG. 3, a computer system 140 receives data from one or more sources. The training data 302 can be received from the The data processor 302 (or a part thereof) may include a user device of an associated user, a volatile substance collection device, a robotic device, a NIR device, one or more other computing devices, and / or one In some embodiments, the computer The data system 140 may select one or more types of training data 302 to receive. It is possible.

[0094]

[0100] The training data 302 includes non-invasive measurements 304, invasive measurements 306, 308, and one or more of the known plant information 308 and the positive infection identification information 310. Non-invasive measurements 304 may be taken using NIR devices, electronic noses, and other volatile collection devices. Non-invasive measurements 304 can be collected using devices such as This may include data showing the volatile composition of the plant as it is emitted from the plant. The actual measurements 304 are taken from plant products on plants in the field (e.g., farm). Thus, the plant product does not have to be harvested from a plant. In this manner, the non-invasive measurements 304 are taken of the plant product as it is harvested / picked from the plant. These can be measured from samples and used for further analysis in a laboratory or other analytical setting. The target measurements 304 may also include a nucleic acid sample of the plant product being analyzed.

[0095]

[0101] Invasive measurements 306 may be collected using devices such as penetrometers and hardness meters. The invasive measurements 306 can be taken by an associated user while the plant product is still part of the plant. collected from plants and / or harvested from plants and used for further analysis. For example, a relevant user may purchase a quantity of plant products (e.g., fruit) from a plant (e.g., tree) and bring that quantity of plant product back to a laboratory or other analytical setting. In the laboratory, the relevant user can, for example, squeeze the plant product to identify the hardness measurement, measuring the thickness of the skin by puncturing the plant product, etc. Invasive measurements can capture measurements of plant products such as dry matter percentage, fatty acid percentage, etc. , sugar content / sugar profiling, and / or acidity / acid profiling. Furthermore, such information can be obtained by volatile analysis and metabolomics. Various other types of analytical methods can be used to identify, but are not limited to,

[0096]

[0102] Known plant information 308 includes growth history information about a particular plant, such as the length of its growing season. This may include information about the environmental conditions that support the growth, maturation, and / or ripening of a particular plant. Environmental conditions can include rainfall levels, sunlight levels, and / or the growth, maturation, and ripening of specific plants. can include a preferred temperature.

[0097]

[0103] Infection positive identification 310 indicates whether a particular plant has a latent infection or is otherwise It may be possible to indicate cases where a person has been identified by law as having a developing latent infection. For example, identifying information 310 is a database of biomarkers that are positively associated with one or more types of latent infection in specific plants. These associations can be made by the associated users. These associations may also be performed by a computer, such as computer system 140. The positive infection identification information 310 may include information such as biomarker levels or expression, environmental conditions, and specific One or more other tests for plants positively associated with one or more types of latent infection in plants. A combination of information can also be shown.

[0098]

[0104] Once the computer system 140 receives the training data (step A ), the computer system 140 correlates the training data based on the predictive features. In other words, the computer system 140 can perform non-invasive Measurements 304, invasive measurements 306, known plant information 308, and / or positive infection identification information The association can indicate the likelihood of latent infection of a particular plant. For example, specific volatile compounds emitted by specific plant products and abnormalities during the growing season can be identified. Rainfall patterns, in combination with temperature, affect the quality of that particular plant and its similar origins during the growing season before harvest. This combination of information can indicate the development of latent infection in a plant. It can be correlated with other information 310.

[0099]

[0105] As mentioned, the training data can be correlated based on the predictive features. Prediction features include plant type, growing area, growing stage, ripening conditions, ripening stage, and environmental conditions. etc. Thus, the predictive features may include, but are not limited to, the onset of latent infection. The growth period, maturation period, and / or maturation period of specific plants and plants of similar origin that may be affected by the disease. Different characteristics during ripening can be represented. In some embodiments, the model for each predictive characteristic For example, you can generate avocados based on where they were grown. As another example, a model can be generated to predict latent infection during the developmental stage. Based on the volatile compounds emitted by certain types of avocados, Predict latent infection (e.g., a specific type of latent infection, such as SER or any type of latent infection) Another model can be generated that allows for the pre-harvest collection of plant gene expression data. Based on the results, other models can be generated to predict latent infection in specific types of produce. For example, one model predicts latent infection in avocados during the first 3 months of growth. Another model can be used to study latent infection in avocados during the first 3–6 months of growth. Gene expression changes based on the age of the plant product. Therefore, gene expression was collected and analyzed during the first 3 months of development. provide different indications of latent infection based on whether the infection is present during the first 3 to 6 months of life or during the first 3 to 6 months of life. This can be beneficial because it allows for the generation of a model using one or more Other time intervals may also be used.

[0100]

[0106] In some embodiments, one or more of the predictive features are combined to generate a model. For example, a combination of volatile compounds emitted by avocados and environmental conditions can Based on this, a model can be generated to predict latent infection in avocados. For example, a combination of environmental conditions (e.g., temperature, rainfall), lime growth stage, and growing region may be used. Based on this, another model can be developed to predict such latent Lyme infection. Other combinations of one or more of the measurement characteristics are also possible. Latent infection of different types of plants based on region, emitted volatiles, and environmental conditions So we can use the same model to generate a model that predicts the size of avocados. The latent aroma of different plants, including but not limited to limes, lemons, apples, citrus fruits, etc. It is possible to predict the stain.

[0101]

[0107] Additionally, in some embodiments, a model can be generated for each growth region. Then, additional models are generated for each region based on age, life stage, environmental conditions, and other information. Therefore, a more accurate prediction of latent infection in plants can be achieved early, before harvest time. You can create multiple models within multiple models to provide a measurement. and mitigation actions early enough before harvest to reduce or otherwise eliminate wastage of plant material at harvest. can be determined and implemented.

[0102]

[0108] The computer system 140 uses the correlated data to estimate the likelihood of infection. A model can be trained to predict (Step C). For example, The received data can be provided as input to the model. Generate output as a value with the likelihood of infection for a particular plant associated with the correlated data. The computer system 140 can convert the output of the model into positive infection identification information 31. Compared to 0, it can further refine and improve the accuracy of the predictive decisions made by the model. Once the desired level of accuracy is reached, the computer system 140 outputs the model. (Step D). The computer system 140 may then execute the model evolution during runtime. ,As it is used, a feedback loop can ,also continually improve the model.

[0103]

[0109] Overall, several biomarkers of plant products can be used to determine the production and / or can determine the model to be generated. The model was used to predict the likelihood of developing latent infection before harvest using the biomarkers identified. Accuracy can be verified. Hormones, gene expression, dry matter (e.g. oil content, maturity), etc. to be generated and / or used during runtime using invasive and / or non-invasive collection of The model can also be determined.

[0104]

[0110] The model is based on the assumption that a particular plant and / or plants of similar origin can harbor a particular type of latent infection or Output showing the likelihood of developing or currently developing any type of latent infection In some implementations, the model can be trained to generate a predictive One or more actions that the relevant user can take to prevent the onset of the observed latent infection. It can also be trained to determine mitigating actions. For example, the model can Select one or more mitigating actions from a repository of known mitigating actions that users typically perform. The model can be trained to reduce or eliminate plant waste at harvest. Users can be trained to select the best mitigation actions to prevent injuries. Users need to think about what kind of mitigating action might be preferable in a particular situation. This can save the relevant user time because there is no need to consider suggested mitigating actions and select one or more such mitigating actions to automatically or It can be performed semi-automatically.

[0105]

[0111] In some embodiments, the model is generated using classification-based hierarchical cluster analysis. You can build and train a model, for example using k-means. Clustering analysis can reveal that gene expression may have similar trends, and therefore This can be beneficial as it can provide clearer clusters for analysis and predictive decisions. Therefore, the model can predict the onset of infection more accurately. training involves one or more methods, including but not limited to random forest modeling. can also use several other techniques: gene expression data, plant information, and other invasion Use decision trees to correlate / associate invasive and / or non-invasive plant measurements / data Therefore, any of the techniques described herein and incorporated by reference may be used. 304 to correlate non-invasive measurements with invasive measurements to predict pre-harvest onset of infection. In some implementations, the model can be trained to determine The training and use of the method does not require invasive measurements 306, but instead involves measuring the composition of volatile components. Pre-harvest incubation using non-invasive measurements304 of maturity, gene expression, and other biomarkers It can be relied upon to predict the onset of infection.

[0106]

[0112] FIG. 4 is a flow chart of a process 400 for predicting the likelihood of infection in a plant. Process 400 predicts the onset of latent infection in a particular plant and / or plants of similar origin. Such predictions can be made at any time before harvest, six months before harvest. The process 400 can be used to reduce or prevent waste of plant material before harvest. Harvesting of plants and / or plants of similar origin so that mitigation actions can be carried out well in advance It can also be performed at any other time before.

[0107]

[0113] Process 400 is illustrated and described throughout this disclosure (e.g., FIG. 1A, 5) can be executed by computer system 140. 0 and / or one or more blocks of process 400 may be implemented in other computing systems, devices, , networks of computers / devices, cloud-based systems, and / or cloud For illustrative purposes, process 400 is explained from the perspective of a computer system.

[0108]

[0114] Referring to process 400 of FIG. 4, the computer system Further considerations regarding receiving plant data: See Figures 1 and 2. At 404, the computer system processes the plant data. In other words, the analysis of plant data can identify predictive characteristics of the plant based on the can be used to select an appropriate model to predict the likelihood of infection of plants This can lead to the identification of one or more characteristics that are relevant to the plant species, growth stage, geographic region, etc. These may include area / location, soil type, age, hormones, gene expression, dry matter, and environmental conditions. As described herein, each plant species, growth stage, geographical area, differing in area / location, soil type, age, hormones, gene expression, dry matter, and / or environmental conditions. Therefore, predictive models can be used to identify predictive features based on plant data. By distinguishing between the two, the computer system is able to predict latent infection in the corresponding plants. Sometimes, plant data can be used to determine the appropriate number of predictive models and the appropriate predictive models. Using the geographical area / location features identified from A subset of models can be selected to be trained to predict infection. These also include environmental conditions or one or more predictive characteristics identified based on plant data. Select one or more models from a subset of models selected based on geographic region / location characteristics using Multiple models can be selected. Therefore, the computer system In this case, one or more models to apply can be identified based on the predictive features.

[0109]

[0115] At 408, the computer system may obtain the identified model. The computer system then applies the model to the plant data at 410. Plant data can be provided as input to the model. One or more different parts of can be input to each model. For example, one model , environmental factors as inputs for predicting the likelihood of latent infection and / or the likelihood of specific types of latent infection. Another model can be used to estimate the likelihood of latent infection and / or the likelihood of a particular type of latent infection. Plant gene expression, volatile composition, and / or hardness as inputs for predicting infection likelihood Measurements can be received from plant data. Yet another model is the likelihood of a plant being latently infected. It can receive a combination of different inputs, such as environmental conditions and gene expression, to predict The model can be trained to receive one or more other inputs and / or combinations of inputs. You can train.

[0110]

[0116] The computer system, at 412, performs a computation based on the application of the model to the plant data. For example, a computer system can identify the pre-harvest quality of a plant based on the The outputs from one or more models, alone and / or in combination, fall within some predetermined threshold range. If the output exceeds a predetermined threshold range or value, In this case, the computer system determines whether the plant will develop a latent infection and / or whether the plant will develop a latent infection. On the other hand, if the output exceeds a predetermined threshold range or value, If not, the computer system determines whether the plant is to determine that they do not have any latent infection and / or are unlikely to develop a latent infection can.

[0111]

[0117] As another example, the computer system may calculate the average of the outputs from the applied model. and determine whether the average power exceeds a predetermined threshold range or value. The computer system may also identify the arithmetic mean output of the model. In this case, the computer system implements majority rule and the majority of applied models The computer system may then identify the output represented in 412. The output can be used as a pre-harvest quality of the plant. The computer system can identify the pre-harvest quality of plants of similar origin. The object may be a tree, field, or other growing area of ​​the same plant for which plant data was received in 402. In some embodiments, the same Plants of similar origin may be in different growing areas, but the plant data is received at 402. The plant may have similar environmental conditions, soil, and / or other predicted characteristics to the plant.

[0112]

[0118] 414, the computer system determines the specific pre-harvest qualities of the plant. The computer system can generate an output for the identified pre-harvest The quality can also be stored. The computer system sends the output to the user device. The output can also be distributed to other users, such as farmers or other users in the supply chain. The output is presented to relevant users. The output is used to assess the pre-harvest quality of the plants and / or the time the plants will be harvested. The output can indicate the pre-harvest quality of plants of similar origin. The computer system can also be implemented to mitigate pre-harvest onset of latent infection. Optionally, the computer may also suggest one or more actions that can be taken. The system may automatically take one or more mitigating actions. The mitigation actions may include any of the actions listed in Table 1 above. One or more other mitigations Actions may also be suggested and / or performed.

[0113]

[0119] FIG. 5 illustrates one or more stimuli that can be used to implement the techniques described herein. 1 is a system diagram of several components of a computer system as described herein. 140, user devices 110A-N, plant data store 230, and model data store The collection device 500 can communicate with the network 150. , can communicate with the components described herein via a network 150. The collection device 500 may be a handheld sample collection device, a sample collection robot, or the like. robot device 120, harvesting robot devices 160, 162, 164, electronic nose / smell detector and / or other automated robots capable of collecting samples from plants and plant products. 1A, particularly those described herein, including but not limited to, Any collection device may be included.

[0114]

[0120] The computer system 140 includes a model training module 502, a data data processing engine 504, latent infection prediction engine 153, latent infection detection module 154, The system may include an abatement engine 155, an output generator 506, and a communication interface 508. The computer system 140 can optionally implement the techniques described throughout this disclosure. one or more additional components, fewer components, or can include different components.

[0115]

[0121] The model training engine 502 trains the predictive models 510A-N. The engine 502 can be configured to receive training data. The training data can be stored in the plant data store 230 and / or the model data store 230. 240. The engine 502 uses the training data to Prediction models 510A-N to predict preharvest latent infection onset in B. typhimurium The engine 502 can analyze the growth area, plant type, plant growth stage, and / or environmental conditions. The engine 502 can train predictive models 510A-N based on the conditions. , as described throughout this disclosure, one or more additional, fewer, or different Furthermore, the engine 502 can train predictive models 510A-N. It is also possible to train multiple models within multiple models. The predictive decisions made during runtime are used to continuously train and / or The model generated and trained by the engine 502 can be used to validate the model. The predictive models 510A-N can be stored in the model data store 240.

[0116]

[0122] The data processing engine 504 processes the collected data as described throughout this disclosure. The device 500 may be configured to process plant data received from the In an embodiment, the plant data collected by the collection device 500 includes plant information 512A- N in the plant data store 230. The engine 504 then Retrieve the plant information 512A-N from the plant data store 230 and analyze it as described herein. Plant information 512A-N can process information such as infection prediction, growing area, environmental conditions, etc. , group stage, non-invasive measurements, invasive measurements, field, plow, or other growing location The data may include, but is not limited to, growth zones in the I can't.

[0117]

[0123] The biomarker expression engine 151 is configured to perform the techniques described in FIG. 1A. The input generation engine 152 can also be configured to perform the techniques described in FIG. 1A. It can be configured so that

[0118]

[0124] The latent infection prediction engine 153 is configured to predict whether a plant and / or plants of similar origin are latent before harvest. The present invention is directed to predicting the likelihood of developing, having developed, and / or being likely to develop a latent infection. For further discussion of the latent infection prediction engine 153, see , see FIG. 1A. Briefly, the latent infection prediction engine 153 is configured to collect data from a collection device 5 00, plant data store 230, and / or data processing engine 504 Using the plant data, the latent infection prediction engine 153 can generate a model From the data store 240, one or more of the predictive models 510A-N to be applied to the plant data are selected. Multiple selections can be made. Based on the application of the selected model 510A-N, The infection prediction engine 153 predicts whether plants are developing, may develop, or are at risk of developing latent infection before harvest. The latent infection prediction engine 153 can predict the likelihood of the onset of the disease and / or the likelihood of the onset of the disease. This prediction can be stored in the plant data store 230 as part of the information 512A-N. .

[0119]

[0125] The latent infection detection module 154 detects latent infections at pre-harvest stages as described with reference to FIG. 1A. determining one or more mitigation efforts that can be implemented in response to the predicted likelihood of latent infection of the Further discussion of the latent infection detection module 154 follows. See Figure 1A for details.

[0120]

[0126] The mitigation engine 155 is adapted to detect latent infections in plants and / or plants of similar origin prior to harvest. Whether one or more actions can be taken to mitigate the potential onset; and The engine 155 may be configured to determine whether or not to execute the The engine 155 may also determine one or more suggestions to take action. Receive predictions from the infection prediction engine 153 (or plant information 51 in the plant data store 230) 2A-N) to determine one or more mitigation actions. In some embodiments, the mitigation engine 155 optionally sends a The system may take one or more of the mitigating actions with or without approval from the user. For example: The plant and / or plants of similar origin have already developed a latent infection and the plant system is dead by the time of harvest. In situations where the infection is likely to continue to develop to the point of failure, the mitigation engine 155 One or more mitigation actions can be automatically implemented to attempt and reduce the continued development of an infection. Cut.

[0121]

[0127] The output generator 506 generates the output to be presented to the user devices 110A-N. The output generator 506 receives the predicted value from the latent infection prediction engine 153. The mitigation engine 155 may receive a measurement decision and / or a mitigation action from the mitigation engine 155. Using this information, the output generator 506 generates messages, notifications, alerts, and / or advertisements. The generated output may be, for example, a plant and / or The output generated may include a value indicating the predicted likelihood of latent infection in plants of the same origin. and / or automatically) selectable and / or implemented mitigation actions. The output may also include one or more other messages as described herein. , notifications, alerts, and / or alarms. The output generator 506 generates the output can be transmitted to the user devices 110A-N for presentation to the users. The generator 506 sends the output to the plant data store 230 for storage in the plant information 512A-N. You can also send it to.

[0122]

[0128] Finally, the communication interface 508 is one of the components described herein. Or communication can be provided between multiple devices.

[0123]

[0129] User devices 110A-N are computers, laptops, and other devices that can be used by associated users. any one or more of a laptop, tablet, mobile device, smartphone, mobile phone, etc. The user devices 110A to 110N may include an input device, an output device, and the like. Such components may include: A user provides information to the computer system 140 and executes the operations of the components described herein. View and / or access information from and / or provided to one or more to be able to act based on this.

[0124]

[0130] 6A-6C are examples of collecting plant data for use in conjunction with the disclosed techniques. 6-10 show how the disclosed techniques can be applied to certain types of plants. 6 to 10 are merely examples, and are provided to illustrate how the present invention can be used. Other perennial fruiting trees (e.g., mango, citrus, pear, apple, and stone fruit trees), grapes (e.g., grapevines), and shrubs (e.g., blackberry bushes, raspberry bushes, and / or limit the application of the disclosed techniques to other types of plants, including raspberry bushes (or strawberry bushes). do not have.

[0125]

[0131] Referring to Figures 6 to 10, an avocado tree (Persea americana) ricana)) can produce large quantities of highly desirable fruit crops for decades. It can be a long-lived perennial plant. Throughout its lifespan, the avocado tree They are constantly interacting with and able to respond to pathogenic bacteria. They may be commensal or mutualistic, but have serious effects on tree health, fruit yield, and fruit quality. There are also many cases of parasites of avocado trees and / or fruit that can cause significant losses. Stem rot (SER) is a disease that occurs near the base of the stem as the fruit ripens after harvest. It is a common quality defect in avocado fruit caused by parasitic infestation of the avocado fruit pulp. The fungus that causes this disease typically produces disease symptoms on the fruit or tree. They can survive parasitically on the pedicels of the fruit before harvest and ripening.

[0126]

[0132] Thus, the disclosed techniques can be used to harvest early, mid, and late avocados. Correlate physical and molecular attributes of fruit with low or high incidence of SER during harvest and ripening. RNA sequencing can be used to generate expression profiles. The expression profiles were obtained from high SER-incidence trees at all time points before peak harvest maturity. These fruits are known to suppress fungal pressure through increased expression of genes involved in the canonical pathogen response pathway. This information can be used to demonstrate that the patient is able to recognize and respond to increased stress. This will be utilized to generate an emergence assay to detect the emergence of SER or other latent infections in avocados. Modifications to fungicide applications, fruit harvest schedules, or fruit market destinations may be made to mitigate the effects. It allows prediction of high and low incidence trees with ample time to perform the correction.

[0127]

[0133] In process 600, 602 is a sample of an avocado. The heat map 604 is an aerial photograph of the trees (see, for example, Figure 6A). The incidence values ​​for each tree at each time point are shown in Figure 6B. The map 604 shows the fruit yields in each box. Furthermore, the blank boxes indicate that a particular tree is producing less fruit. or were not available for sampling at the time due to ongoing tree management / pruning. The image data 606 shows that the number of days after flowering (DAF) is 330 days and the number of days after flowering (t1) is 330 days. The difference between the low (C2) incidence tree and the high (C1) incidence tree at 3 (DAF 520 days) ( See, for example, Figure 6C ).

[0128]

[0134] Avocado fruit can be collected from trees in fields or other growing areas. As shown in aerial photograph 602 in Fig. 6A, the samples were collected from 13 plots in Block 7 of the field. Fruits can be collected periodically throughout the growing season. All fruits The fruit was sprayed frequently with 70% EtOH throughout the harvest, leaving about 2-4 cm of pedicel. Harvest by gently picking the fruit using sterilized stainless steel scissors. In the example process 600, at each collection time, 30 fruits are harvested from each tree. can be harvested from and transported to a research center in a crate (approximately 15 minute transport time) A total of 1,050 fruits can be collected from these trees. As shown in 02 and 704, weather data for the harvest day is also retrieved from the data store. The time points analyzed in the example process 600 are approximately 330 days DAF, approximately 43 days DAF, and approximately 50 days DAF. The time point can be from day 0, about 520 days DAF.

[0129]

[0135] Upon arrival at the research center, the fruit were placed in thin produce crates and left overnight at ambient temperature. Approximately 24 hours after harvest, all samples (each tree at each time point) were The fruit mass (g) and respiration rate (mL CO2 / kg-hr) were measured individually (n=30). Fruit respiration can be measured by placing the fruit in a custom airtight pod equipped with a CO2 sensor. It can be measured by sealing the fruit and measuring the CO2 concentration every 5 seconds. Fruit dry matter content was measured using a near-infrared handheld spectrometer with a custom dry matter prediction algorithm. It is also possible to take images of the external quality of the fruit. At each time point, a subset of fruits (n = 4 per tree per time point) was destructively sampled for RNA extraction. The remaining fruit (n=24) was allowed to ripen under ambient conditions. For the fruit to reach 50 Shore (measured using a handheld hardness tester) The required time (in days) can be considered as the time to ripen. The fruit is then cut in half and inspected for quality defects (including SER), and image data is collected. An image can be taken as shown at 606.

[0130]

[0136] Once all collections were completed, the incidence of SER in ripened fruit on each individual tree at each time point was calculated. The results showed that fruit marked an increase in SER incidence over the season. Trees can be designated as "high incidence," while trees that produce fruit with low SER incidence are The three "high incidence" ( A1, C1, and D1) trees and three "low incidence" (A4, C2, and D2) trees in detail. can be selected for transcriptional analysis, see heatmap 604 in Figure 6B.

[0131]

[0137] As mentioned above, at each time point, 24 hours after harvest, cross sections of the fruit (seeds, endocarp) were The pulp (which may contain the mesocarp, mesocarp, and exocarp) is cut and ground under liquid nitrogen to a fine powder. By cutting the fruit, four fruits can be destructively sampled from each tree. These sections can contain the peel, pulp, and seeds. The samples were used for RNA extraction. It can be stored at -80°C until use.

[0132]

[0138] Individual samples from the "high incidence" and "low incidence" trees collected at each time point were then analyzed. Each sample was purified using bead cleanup with cetyltrimethylammonium bromide (C RNA can be extracted using a buffer-based extraction protocol. Prepare RNA-seq libraries and pool by time point for sequencing. The library can be run in some embodiments at 75 cycles (high output). It can be run to assess file quality and align to a specific avocado reference genome. It can also perform BAM file generation and indexing.

[0133]

[0139] Sequence reads can also be aligned to genomic features. A hand count table can be prepared. At each time point, the difference between the high and low incidence trees is Only genes with low counts (>10 counts) were retained. Differentially expressed genes were filtered for p-values ​​less than 0.05. can.

[0134]

[0140] Perform principal component analysis and visualize it on the transformed normalized DeSeq matrix. In preparation for the heatmap visualization, you can also Fold change matrix of regulated genes (filtered for p-values ​​less than 0.05) The risk can be normalized to the value of each gene across samples using the R function hchust. Genes can be clustered by their expression patterns within each time point using In some embodiments, the R function pheatmap is used to Visualize how individual samples cluster based on gene expression patterns can be done.

[0135]

[0141] Perform and visualize clustering of differentially expressed genes by time-course expression trend Such a function can calculate gene similarity over time across samples within a condition. Relative gene expression can be normalized to a Z-score, where the values ​​are centered around the arithmetic mean of each gene. and scaled to the standard deviation for each gene. Genes are, in some embodiments, included in the top 500 upregulations with p-values ​​less than 0.05. The top 500 differentially expressed genes and the top 500 down-regulated differentially expressed genes were analyzed. The gene clusters examined can be limited to those with gene counts greater than 20. can be limited to.

[0136]

[0142] Using the designated homologous gene in Arabidopsis for each avocado transcript, For the list of top differentially regulated avocado genes by points and expression clusters Gene ontology analysis can be performed on these lists. Enrichment analysis that can map gene lists to known sources of functional information Used as input to the tool to identify statistically significant enriched biological processes and pathways , regulatory motifs, and protein complexes can be detected. The adjusted enrichment p-values ​​of the classes can be identified and output (eg, reported).

[0137]

[0143] Overall, Figures 6A-6C show the locations (e.g., areas) of sampled trees. The SER incidence rate of fruit collected from each tree at each time point is shown. At 330 days after harvest, no SER was observed. Over the next two harvests, SER occurrence The rate increased on average, with the most significant increase in fruit from trees A1, C1, and D1. Samples from these trees can then be assigned to the "high incidence" group. Nearby trees A4, C2, and D2, which have minimal or no SER, were labeled as “low-risk” trees as control comparisons. can be assigned to a "survival rate" group.

[0138]

[0144] 7A and 7B show the physical distribution of the plants in FIG. 6 that can be harvested at different times. Graph 702 in FIG. 7A illustrates a graphical representation of example attributes of the fruit collection system described herein. Plot of maximum and minimum temperatures across the body, with sampling times 1, 2, and Graph 704 in FIG. 7A shows the overall fruit harvest described herein. The plot shows the rainfall measured over the area, with sample times marked 1, 2, and 3. Graph 706 in FIG. 7B shows the results for DAF 330 days, DAF 430 days, and DAF 52 days. Graph 708 in Figure 7B shows the average initial respiration per fruit collected on day 0. The mean mass of fruit collected at each time point from the high and low SER incidence trees is shown, with the standard deviation being the error bar. The graph is plotted as

[0139]

[0145] Referring to both Figures 7A and 7B, the respiration rate 24 hours after harvest was The maximum fruit yield from all trees was achieved at the beginning of February, which was 330 days after AF ( For example, see graph 706 in Figure 7B.) This is due to the fact that temperatures are lower on average at the beginning of the year (for example, see Figure 7B). 7A, graph 702), and / or rapid growth and / or cell division leads to increased metabolism, This may be the stage of fruit development at that time, as increased respiration may be required. No significant differences in mean dry matter content or respiration rate were observed between high and low SER incidence trees. Although it may not be possible to observe a trend between the average mass of individual fruits and the incidence of SER, As the number of days after flowering increases, fruit mass can also increase (e.g., Fig. 7B, graph 708), but mean fruit mass was consistently higher in fruit from the high SER incidence group. This trend is most pronounced in the third collection (520 days DAF), and In the first collection, the average fruit mass from A1, C1, and D1 was greater than 250 g. , while the average fruit mass from A4, C2, and D2 is all less than 200 g (e.g., Fig. See graph 708 in 7B).

[0140]

[0146] 8A-8C show the results of the data collection of the plants in FIGS. 6A-6C over the entire period of time. A biplot of the principal component analysis (PCA) of normalized gene expression is shown. The 800 data set was used to analyze gene expression data at 330 days after administration of SER in both the low and high SER incidence groups. The biplot 802 shows the first two principal components from the PCA of the data. The first two main results from PCA of gene expression data at 430 days DAF for both survival groups Biplot 804 shows the components of the 520-day DAF in both the low and high SER incidence groups. Figure 1 shows the first two principal components from a PCA of gene expression data in

[0141]

[0147] Referring to biplots 800-804 in Figures 8A-8C, all three sequences Over the course of a sequencing run, 5-6 million reads can be obtained per sample. The percentage of identified leads (%RF) and the percentage of Q scores above 30 (%≥Q30) were The percentages are over 90% and 95%, respectively. After alignment to the (Americana cv. Hass) genome, the reads averaged 81.4% and 80.8%. %, and 80.9% were observed at t1 (DAF 330 days), t2 (DAF 430 days), and t3 ( The data set was successfully assigned to the annotation regions of the genome. It can be done.

[0142]

[0148] Visualization of PCA of normalized expression across all time points using pair plots It can be shown that the strongest separation by PCA occurs between time points, especially PC1-PC2. There is some slight separation between the low and high SER groups in PC3 and PC4. However, if PCA is performed within each individual time point and biplots are used, When visualized as a low-order α-glucan complex, low-order α-glucan complexes were observed across PC1 and PC2, as shown in Figures 8A to 8C. A slightly clearer separation between the SER and high SER samples can be observed. This separation is most evident at the latest time point, 520 days DAF. .

[0143]

[0149] 9A-9C show the normalized gene expression profiles of the plants in FIGS. 6A-6C. The volcano plot in Figure 9A shows the low-energy volcano at 330 days DAF. Volcano plot in Figure 9B showing differential gene expression between the live and high incidence tree groups. 902 Differential gene expression between low and high incidence tree groups at 430 days DAF The volcano plot 904 in Figure 9C shows the low incidence and high incidence tree groups at 520 days DAF. Plots 900-904 show differential gene expression between incidence tree groups. Normalized gene expression (Log2 fold change) vs statistical significance (-Log 10 In each of the plots 900 to 904, the Log2 phase is greater than |2|. Differentially expressed genes with field changes and p-values ​​less than 0.05 are grouped into quadrants with patterns. The numbers in these quadrants represent the number of genes that meet these eligibility criteria. In Figure 9A, seven genes met the eligibility criteria for quadrant 1, and 11 genes met the eligibility criteria for quadrant 2. The eligibility criteria for the 25 genes and 12 genes in Figure 9B and Figure 9C were met, respectively. In 9C, 8 genes and 22 genes met the eligibility criteria for quadrant 1 and quadrant 2, respectively. satisfy.

[0144]

[0150] Referring to Figures 9A to 9C, at all three time points, the high incidence tree groups (A1, C1 , and D1) compared with the low incidence tree groups (A4, C2, and D2) revealed differences in gene expression. At t1, 863 genes were upregulated in the high incidence tree. We were able to downregulate 963 genes (4% of the identified genes). At t2, 101 genes were identified in the high-incidence tree (4.4% of the identified genes). Two genes could be upregulated (4.8% of identified genes), and 7 86 genes can be downregulated (3.7% of identified genes) At t3, 234 genes were upregulated in the high incidence tree ( 1% of the identified genes, 166 genes can be downregulated ( 0.08% of identified genes. These DEGs had p-values ​​less than 0.05 and p-values ​​less than -2. or if filtering for more than two fold changes, then The number of up- and down-regulated genes in the high incidence tree was 11 up- and 7 down-regulated, respectively. , 12 up and 25 down, and 22 up and 8 down.

[0145]

[0151] In some embodiments, the top 30 at each time point are up- and down-regulated. Genes (pre-filtered for p-values ​​less than 0.05 prior to sorting by fold change) A heatmap visualization (not shown) of the high incidence of genotypes from the trees marked as high incidence is shown. This indicates that the individual fruit expression patterns of DAF33 cluster together. These genetic changes in fruit were observed as early as the first day of February, which was day 0, and as early as six months before the later harvest season. These trees can be assigned high or low incidence based on the gene expression patterns of the offspring. In particular, there is little overlap between these genes across the three time points. Only four genes can be conserved across these three groups: mak er-Ctg1243-augustus-gene-0.16-mRNA-1, mak er-Ctg0870-augustus-gene-0.10-mRNA-1, aug ustus_masked-Ctg0285-processed-gene-0.0- mRNA-1, and maker-Ctg0255-augustus-gene-4.1 6-mRNA-1. The predicted closest mRNA for each of these genes is from tomato (S. lycopersicum) (S. lycopersicum) and Arabidopsis (A. thaliana) homologs These results can be discerned by transcriptome-level analysis. changes in response to latent fungal pathogenic mechanisms in the fruit throughout the fruit development phase on the tree. This may suggest that:

[0146]

[0152] Statistical enrichment of the top 30 up- and down-regulated genes at each time point The chromosome analysis can use a list of Arabidopsis homologs as input. The results were based on gene ontology sequences with a p-value of 0.05 or less and a false discovery rate (FDR) of less than 1. In the example of FIGS. 6A to 6C, t1, t2 At time points t1 and t2, 20, 35, and 79 genes met these criteria, respectively. There can be gene ontology terms. This analysis is done by dividing the low-SER tree and the high-SER tree. The biological processes most significantly represented in this subset of transcripts may show significant differences between It can provide the highest level of insight into molecular function and cellular component classification.

[0147]

[0153] Identification of the top 30 up- and down-regulated gene sets across time points As expected given the limited overlap of genes, the most significant Gene Ontology classifications The overlap between t1 and t2 may be limited, and the overlap between t1 and t3 may be greater than the overlap between t1 and t3 and between t2 and t3. For example, the significant GO datasets for t1 and t2 both contain the biological The process terms response to other organisms (GO:0051707) and response to external biological stimuli (GO:0051708) are also included. Response (GO:0043207) was analyzed using the cell component term cell margin (GO:007194 4) and the molecular function term transporter activity (GO:0005215). It is possible that none of these terms overlap with t3. In addition, t2 also contains oxidoreductase activity (GO) that may not be present in the t1 dataset. :0016491), response to oxidative stress (GO:0006979), plant organ development ( GO:0099402), response to osmotic stress (GO:0006970), other organic Defense responses to the body (GO:0098542), and aromatic compound biosynthesis (GO:001 The significant GO list for t3 also includes additional terms related to acid. The term for hydroxylase reductase activity (GO:0016491) and the term for aromatic compound biosynthesis (G O:0019438) and can overlap with t2. Related terms for embryonic development (GO:0099402), such as post-embryonic development (GO:0009791) and implantation May include reproductive structure development (GO:0048608). Most likely, t3 overlaps with t1 or t2. Instead, the GO list for t3 is based on responses to hormones (GO:000). 9725), cellular response to hormone stimulation (GO:0032870), and hormone-mediated signal transduction Targets related to hormone signaling, including the signal transduction pathway (GO:0009755) Additionally, the t3 dataset may contain additional data that are not found in the t1 or t2 datasets. These include many GO terms related to transcriptional regulation that are not yet fully understood. O:0006351), transcriptional regulation (GO:0006355), DNA binding (GO:000 3677), RNA biogenesis process (GO:0032774), and RNA biogenesis process regulation ( GO:2001141).

[0148]

[0154] Figures 10A-10D show strong time and condition trends for the plants in Figures 6A-6C. Figure 10 shows normalized gene expression over time for selected gene clusters with Referring to all of Figures 10A through 10D, the numbers 1, 2, and 3 on the x-axis represent DAF 330 days, D The time points of each sample taken on 430 days AF and 520 days DAF are shown. Clustering of differentially expressed genes by NMR is performed and visualized using known computational techniques. Such techniques can measure genetic similarity over time across samples within a condition. Relative expression can be normalized to Z-scores, and values ​​can be centered around the arithmetic mean. The data can be analyzed for trends over time and scaled to the standard deviation of each gene. The genes included were the top 500 up-regulated and The results can be limited to the top 500 down-regulated differentially expressed genes. The gene clusters that are analyzed can be limited to those with gene counts greater than 20. Sixteen gene clusters met these criteria and are shown in Figures 10A-10D. P. americana genes classified into each cluster using GOSt A statistical enrichment analysis was performed using a list of Arabidopsis homologs for The top GO categories for each group are shown in Table 2.

[0149] [Table 6]

[0150] [Table 7]

[0151]

[0155] In particular, as shown in Figures 10A to 10D and Table 2, the highly infected trees were Several groups (groups 9, 11, and 1) showed an increasing trend and had relatively high gene content in the fruit. There may be groups 2 and 18. Genetic studies of these groups for pathways specific to plant pathogen responses have been performed. The GO annotations and the related genes can be examined. Group 11 shows significant genes related to these pathways. For example, it can be found that the number of terms in group 11 is greater than 0.05. Several top biological process GO categories with much smaller p-values ​​were related to responses to stimuli ( GO:0050896), response to chemicals (GO:0042221), and other organisms (GO:0051707), and genes from these categories Examples include predicted endochitinase 4 (m), plotted in Figures 10A-10D. aker-Ctg0022-augustus-gene-3.7), pathogenesis-related protein Protein PR1 precursor (augustus_masked-Ctg0281-processe d-gene-0.6), and osmotin-like pathogenesis-related protein R (augustus_ There is a masked-Ctg2009-processed-gene-0.2) Group 18 further includes, but is not limited to, reactive oxygen species metabolic processes (GO:0072593), Response to stress and regulation of superoxide radical scavenging (GO:2000121) Additionally, group 12 includes GO groups related to the response to jasmonic acid (GO:00 09753), and This can be established as being involved in plant defense signaling. Important differences in gene expression may exist between high-SER and low-SER tree populations, resulting in distinctive These findings could be responsible for shifts in pathogen response pathways.

[0152]

[0156] Genes that showed lower expression in the high SER samples compared to the low SER group (Group 1) In most of the groups (groups 1, 2, and 4), there was a general overrepresentation of genes involved in cell development and differentiation. For example, the most significant biological process GO term in group 1 is floral organ identity. Specification of plant organ identity (GO: 0090701), while the top two in Group 4 were xylem and phloem patterns. These findings are likely to be due to the following reasons: formation (GO:0010051) and regionalization (GO:0003002) This suggests that fruit from trees with higher SER may have different developmental patterns. Furthermore, group 5 is involved in the phenylpropanoid biosynthesis process (GO :0009699). Phenylepropanoid compounds are highly expressed in plants. It is known to be involved in defense against bacteria. Therefore, it is known to be involved in phenylpropanoid biosynthesis. Higher expression of the enzyme can reduce the susceptibility of low-SER fruits to SER infection. In addition, the phenylpropanoid pathway can generate precursors for lignin biosynthesis. Therefore, it may be related to the developmental gene expression differences seen in groups 1, 2, and 4.

[0153]

[0157] Figures 6 to 10 show that differential gene expression analysis reveals differences between high and low SER incidence. Significantly different up- and down-regulated genes in fruit samples from trees grouped according to These trees show that the genetic makeup can be revealed at harvest maturity. Fruit can be classified into high or low SER incidence groups based on the occurrence of SER symptoms in the fruit. Low-maturity fruits (less than about 30% dry matter) were harvested at the first two time points and allowed to ripen. If the SER development is more severe, there may be minimal, if any, SER development, whereas the more mature SER development at the end It can be observed that fruits can develop significant levels of SER. This may be due to the rise in average temperatures leading to the end of the disease. Gene expression data suggest that this may be due to the onset of symptoms. The transcriptome profile of fruit from high-incidence trees can be measured even at these early time points. This could show significant differences in the prevalence of latent infection, facilitating pre-harvest prediction of latent infection.

[0154]

[0158] Furthermore, as described herein in connection with Figures 6-10, most infected trees The specific genes and processes that are significantly altered may shift over the time course of fruit development This is the top three clustering of fruit expression trends into expected high and low SER groups. The overlap between the top 0 up-regulated genes and the top 30 down-regulated genes was the highest. The analysis of gene groups shows that they differ over time. This allows for the relative abundance of the gene across the three sampled time points. These grouped expression profiles may also show dynamic shifts in transcript abundance. Gene ontology analysis of time trends revealed several differentially expressed genes over time in the high SER tree population. Osmotin-like proteins and chitinases that may show a tendency to maintain and / or increase This allows for the isolation of a group of known pathogen response genes, including the jasmine gene. Identifying genes involved in monophosphate signaling and reactive oxygen species metabolism to identify high-incidence individuals at early time points It can be upregulated in the fruit of the tree (e.g., Figures 10A-10D, Table 2). (See Groups 12 and 18 in the previous section.) Reactive oxygen species and jasmonate signaling events may be a known mechanism deployed in the early host defense response against fungal pathogens. This suggests that fruit from trees with a high incidence of SER may actually be more susceptible to infection well before the onset of symptoms. suggest that SER may exhibit a high response to the presence of fungal pathogens. can.

[0155]

[0159] The average incidence of SER after harvest was a maximum of 17% for tree C1, even for high incidence trees. Using four individual fruit samples per tree, it can be observed that The likelihood that these particular fruits will develop SER symptoms is low. The observed genetic differences between fruit from low-incidence trees and fruit from high-incidence trees in the same sample The differential expression can be interpreted as a systematic response throughout the tree to the presence of SER fungi. This can be done.

[0156]

[0160] Furthermore, analysis of grouped genes by differential expression over time (e.g., Figure 10 A-D of Figure 10 and Table 2) further demonstrate that fruit from high SER incidence trees and those from low SER incidence trees are more susceptible to SER. This may suggest differences in fruit development between high-SER and low-SER trees. The difference between the high SER group and the low SER group may be due to the low expression of genes involved in the development and differentiation of the The only physical difference measured was a small difference in the average mass of fruit harvested across the three time points. (See, e.g., Figure 7B) and higher SER trees produce higher average mass fruit.

[0157]

[0161] FIG. 11 shows a system that can be used to implement a system for detecting latent infection. The computing device 1100 may be a laptop, desktop, , workstations, personal digital assistants, servers, blade servers, mainframes, and It is intended to represent various forms of digital computers, including other suitable computers. The computing system 1150 may be a personal digital assistant, a cellular phone, a smart phone, or other similar device. It is intended to represent various forms of mobile devices, such as computing devices. The 1100 or 1150 device supports Universal Serial Bus (USB) flash drives. The USB flash drive contains the operating system and other The USB flash drive can store the application. or a USB connector that can be inserted into the USB port of another computing device The components shown here, their connections and relationships, and the The functions and features described and / or claimed herein are intended to be examples only. It is not intended to limit the scope of the embodiments of the present invention described herein.

[0158]

[0162] The computing device 1100 includes a processor 1102, a memory 1104, a storage device 1106, and a 108, a high-speed interface 11 connected to a memory 1104 and a high-speed expansion port 1110 08, and a low-speed interface 111 connected to a low-speed bus 1114 and a storage device 1108 12. Components 1102, 1104, 1108, 1108, 1110, and 111 Each of the two is interconnected using various buses, connected to a common motherboard or otherwise. The processor 1102 can be installed as needed. to an external input / output device such as a display 1116 coupled to the interface 1108; The computing device 1100 includes instructions stored in memory 1104 or storage device 1108 for displaying In other embodiments, multiple memories may be used. Multiple processors and / or multiple buses may be used, along with multiple types of memory and It is possible. The plurality of computing devices 1100 may also be, for example, a server bank, a group of blade servers, or They can be connected as a multiprocessor system, with each device performing some of the required operations. provide.

[0159]

[0163] The memory 1104 stores information within the computing device 1100. In some embodiments, the memory 1104 is one or more volatile memory units. The memory 1104 is one or more non-volatile memory units. It may also be another form of computer readable medium, such as a magnetic or optical disk.

[0160]

[0164] Storage device 1108 provides mass storage for computing device 1100. is possible.

[0161]

[0165] In one embodiment, the storage device 1108 may be a floppy disk drive, a hard disk drive, Disk device, optical disk device or tape device, flash memory or or other similar solid-state memory devices, or storage area networks or other The computer-readable medium may be a device, an array of devices, or the like. A computer program product may contain or be written on an information carrier. The computer program product, when executed, can tangibly embody the above-mentioned The information carrier may also include instructions for performing one or more methods, such as: 104, storage device 1108, or memory 1102 on a processor. It is a machine-readable medium.

[0162]

[0166] The high-speed controller 1108 controls the bandwidth-intensive operations of the computing device 1100. The slow controller 1112 manages the less bandwidth-intensive operations. This allocation of functions is merely an example. In one embodiment, the high speed controller 1108 , coupled to memory 1104, for example through a graphics processor or accelerator. 1116 and can accept various expansion cards (not shown). In this embodiment, the low-speed controller is coupled to a high-speed expansion port 1110 that can 1112 is coupled to the storage device 1108 and the low-speed expansion port 1114. supports various communication ports such as USB, Bluetooth, Ethernet, wireless The device may include a keyboard, pointing device, and microphone. / speaker pair, scanner, or switch or router, for example, through a network adapter It may be coupled to one or more input / output devices, such as a networking device The computing device 1100 may be implemented in several different forms, as shown. For example, it may be implemented as a standard server 1120 or This can be implemented multiple times within a cluster of servers. In addition, it can be implemented as a personal computer such as a laptop computer 1122. The method can be implemented on a personal computer.

[0163]

[0167] Alternatively, components from computing device 1100 may be integrated into devices such as device 1150. , and can be combined with other components in a mobile device (not shown). Each such device may include one or more of the computing devices 1100, 1150. The entire system may consist of multiple computing devices 1100, 1150 communicating with each other. can.

[0164]

[0168] The computing device 1100 may be implemented in several different forms, as shown. For example, it may be implemented as a standard server 1120 or It can be implemented multiple times within a group of servers. In addition, the system may be implemented as a personal computer such as a laptop computer 1122. It can be implemented on a computer.

[0165]

[0169] Alternatively, components from computing device 1100 may be integrated into devices such as device 1150. , and can be combined with other components in a mobile device (not shown). Each such device may include one or more of the computing devices 1100, 1150. The entire system may consist of multiple computing devices 1100, 1150 communicating with each other. can.

[0166]

[0170] The computing device 1150 includes, among other components, a processor 1152; Memory 1164, input / output devices such as a display 1154, and a communication interface 1 166, and a transceiver 1168. The device 1150 may include a microdrive or other A storage device such as a storage device may also be provided to provide additional storage. Each of 1150, 1152, 1164, 1166, and 1168 uses a different bus. The components are interconnected by a common motherboard or other suitable means. It is possible.

[0167]

[0171] The processor 1152 operates as a computing device, including instructions stored in memory 1164. The processor can execute instructions in the analog processor 1150. The chip may be implemented as a chipset of chips including a digital processor and a Additionally, the processor may be implemented using any of several architectures. For example, the processor 1110 may be a CISC (Complex Instruction Set Computer) processor, RISC (reduced instruction set computer) processor, or MISC (minimal instruction set computer) The processor may be, for example, a user interface (UI) processor. the device 1150, the applications executed by the device 1150, and the The device 1150 may provide coordination of other components of the device 1150, such as control of wireless communications.

[0168]

[0172] The processor 1152 controls the control interface 1158 and the display 115 4. The display interface 1156 may communicate with the user. The display 1154 is, for example, a TFT (thin film transistor liquid crystal display) display. display, OLED (organic light-emitting diode) display, or other suitable display The display interface 1156 can be a display 11 54 to present graphical and other information to a user. The control interface 1158 receives commands from the user and controls the process. In addition, the processor 1152 can convert them for submission to the processor 1152. 52, thereby allowing other devices to communicate with each other. The device 1150 enables short-range communication with the device.

[0169]

[0173] External interface 1162 may, for example, provide wired communication in some embodiments. In other embodiments, wireless communication may be provided, using multiple interfaces. It can also be used.

[0170]

[0174] The memory 1164 stores information within the computing system 1150. 4 may include one or more computer-readable media, one or more volatile units, or a or may be implemented as one or more of multiple non-volatile units. 174, for example, SIMM (Single In-Line Memory Module) card interface The device 1150 is connected through an expansion interface 1172, which may include an Such expansion memory 1174 may provide additional storage space for the device 11. 50 or record application or other information for device 1150. Specifically, the expansion memory 1174 may also store the above-mentioned processes. It may contain instructions for supplementation and may also contain secure information. For example, expansion memory 1174 may be provided to device 1150 as a security module. and programming instructions that enable secure use of the device 1150. In addition, it places identifying information on the SIMM card so that it cannot be hacked. etc., to provide secure applications with additional information via SIMM cards. can be done.

[0171]

[0175] The memory may be, for example, a flash memory and / or a non-volatile RAM, as described below. In one embodiment, the computer A computer program product may be tangibly embodied in an information carrier. The program product includes instructions that, when executed, perform one or more methods, such as those described above. The information carrier may be the memory 1164, the expansion memory 1174, or the processor. 1152, and the like, and the transceiver 116 8 or external interface 1162.

[0172]

[0176] Device 1150 may include digital signal processing circuitry if necessary. The device can communicate wirelessly through a communication interface 1166. 1166 is used, among other things, to make GSM voice calls, SMS, EMS or MMS messages. Zing, CDMA, TDMA, PDC, WCDMA, CDMA2000, GPRS, etc. Such communications may be provided under a variety of modes or protocols, e.g. For example, this can be done through a radio frequency transceiver 1168. In addition, short-range communication can be Such communication may be performed using Bluetooth, Wi-Fi, or other such transceivers (not shown). In addition, a GPS (Global Positioning System) receiver module 1170 can Additional navigation and location related wireless data may be provided to the device 1150. The navigation and location related wireless data is transmitted to applications running on the device 1150. It can be used appropriately depending on the application.

[0173]

[0177] The device 1150 communicates audibly using an audio codec 1160. The audio codec 1160 receives audio information from the user and can The audio codec 1160 can also convert audio data into digital information that can be used for audio transmission. For example, device 1150 may generate an audible sound for the user through a headset speaker. Such sounds may include sounds from voice telephone calls, recorded sounds, etc. (e.g., voice messages, music files, etc.) running on device 1150. This may also include sounds created by applications that create the sound.

[0174]

[0178] The computing device 1150 may be implemented in several different forms, as shown. For example, the computing device 1150 may be implemented as a cellular telephone 1180. The computing device 1150 may be a smartphone 1182, a personal digital assistant, or other The present invention may also be implemented as part of a similar mobile device.

[0175]

[0179] Various embodiments of the systems and methods described herein may be implemented using digital electronic circuits, Integrated circuits, specially designed ASICs (application-specific integrated circuits), computer hardware Implemented in hardware, firmware, software, and / or combinations of such implementations. These various embodiments may include a storage system, at least one input and at least one output device, and a storage system. a system, at least one input device, and at least one output device to transmit data and instructions; at least one programmer, which may be dedicated or general-purpose, coupled to transmit instructions; one or more programs executable and / or interpretable on a programmable system including a programmable processor may include implementations in multiple computer programs.

[0176]

[0180] These computer programs (programs, software, software An application (also known as code) is a program that executes machine instructions for a programmable processor. Includes high-level procedural programming languages, object-oriented programming languages, and and / or assembly / machine language. The terms "machine-readable medium" and "computer-readable medium" refer to a medium that can be written as a machine-readable signal to carry machine instructions. a programmable processor including a machine-readable medium for receiving machine instructions and / or data; Any computer program product, apparatus, and / or device used to provide devices, such as magnetic disks, optical disks, memories, and programmable logic devices (PLDs) The term "machine-readable signal" refers to a signal that can be used to provide machine instructions and / or information to a programmable processor. refers to any signal used to provide data.

[0177]

[0181] To provide for user interaction, the systems and techniques described herein A display device, such as a CRT (cathode ray tube) or LC D (liquid crystal display) monitor and keys that allow the user to enter input into the computer a computer with a keyboard and a pointing device, such as a mouse or trackball Other types of devices may be used to provide user interaction as well. For example, the feedback provided to the user may be any form of sensory feedback. feedback, such as visual feedback, auditory feedback, or tactile feedback The input from the user can be any input, including acoustic, speech, or tactile input. The information can be received in any form.

[0178]

[0182] The systems and techniques described herein may be implemented in a backend, for example as a data server. or middleware components, such as application servers, or front-end components, such as a user interface, that implement the systems and techniques described herein. A graphical user interface or web browser that allows interaction with the aspects or a client computer having such backend, middleware or any combination of front-end components. The components of the system may communicate digital data in any form or medium, e.g. They can be interconnected by a network. Examples of communication networks include local area networks. LANs, wide area networks ("WANs"), and internetworks. There is the internet.

[0179]

[0183] A computing system can include clients and servers. The computer and the server are generally remote from each other and typically interact through a communication network. The client and server relationship is established by running on each computer and acting as a client to the other. It arises from a computer program that has a client-server relationship.

[0180]

[0184] Although this specification contains many specific implementation details, these are not intended to limit the scope of the disclosed technology. should not be construed as a limitation on the scope of what may be recited in the claims. Rather, they are intended as descriptions of features that may be specific to particular embodiments of the particular technology disclosed. Certain features described herein in terms of separate embodiments should be construed as such. The features may also be implemented in combination in one embodiment, either in part or in whole. In particular, various features that are described with respect to one embodiment may be used separately or optionally in multiple embodiments. Furthermore, features may be used herein in any suitable subcombination. may be described and / or originally claimed as operating in conjunction with Although one or more features from the combinations recited in the claims may be combined, in some cases In this case, the combination may be deleted, and the combination described in the claims may be a subcombination. Also, the operations may be recited in a particular order. However, this does not mean that such actions must be performed in that particular order or sequences to achieve a desired result. should be understood as requiring that the next or all actions be performed. Certain embodiments of the subject matter have been described. Other embodiments are within the scope of the following claims. is located. In certain embodiments, for example, the following items are provided: (Item 1) 1. A method for identifying pre-harvest latent infection in a plant, comprising: obtaining, by a processor, data describing the expression levels of one or more infection biomarkers present in the plant, the infection biomarkers indicative of an infection likelihood of the plant, and the data indicative of a difference between a read sequence of the plant and a reference genome of a healthy plant of the same species as the plant; selecting, by the processor, one or more machine learning models based on the acquired data, the one or more machine learning models having been previously trained using data correlating other data with the identified one or more infection biomarkers of one or more other plants, and generating an output indicative of the likelihood of the plant developing a latent infection before harvest, the one or more machine learning models being trained using a process comprising: inputting into the one or more machine learning models from a training dataset: (i) non-invasive measurements of the one or more other plants; (ii) invasive measurements of the one or more other plants; (iii) known plant information of the one or more other plants; and (iv) positive infection identification information of the one or more other plants; determining a predicted likelihood that the one or more other plants will develop the latent infection before harvest based on inputting (i)-(iv) into the one or more machine learning models; and outputting the one or more machine learning models for runtime use. selecting, generating, by the processor, an output indicative of the likelihood that the plant will develop the latent infection before harvest based on applying the one or more machine learning models to the data; determining, by the processor, that the plant has a latent infection based on the output exceeding a predetermined threshold range; determining, by the processor, one or more treatments that mitigate the latent infection in the plant; outputting, by the processor, an indication that the plant has the latent infection and the one or more determined treatments; A method comprising: (Item 2) 2. The method of claim 1, wherein determining, by the processor, one or more treatments to mitigate the latent infection in the plant includes identifying, for the plant and one or more other plants of similar origin, at least one of: (i) a harvest schedule that is changed to earlier in the growing season; (ii) instructions to apply a predetermined amount of an insecticide before harvest; and (iii) a time when the plant and the one or more plants of similar origin are approaching the end of their respective healthy plant productive lives, wherein the one or more machine learning models are previously trained to identify (i)-(iii) using data from the training dataset. (Item 3) 3. The method according to item 2, wherein the one or more other plants of similar origin include plants in the same zone as the plant. (Item 4) outputting, by the processor, an indication that the plant has the latent infection and the one or more determined treatments; generating an alert message that, when processed by a user device, causes the user device to output an alert notifying a user of the user device to perform one or more of the determined actions; transmitting the generated alert message to the user device; Item 1. The method according to item 1, comprising: (Item 5) determining, by the processor, one or more treatments to mitigate the latent infection in the plant, determining, based on the generated output, that an antimicrobial treatment should be prescribed to one or more other plants of similar origin prior to harvest; The method according to item 1, comprising: (Item 6) determining, by the processor, one or more treatments to mitigate the latent infection in the plant, generating instructions that, when treated by an irrigation controller, cause the irrigation controller to automatically spray a liquid containing the antimicrobial treatment; transmitting the instructions to the irrigation controller; Item 6. The method according to item 5, comprising: (Item 7) determining, by the processor, one or more treatments to mitigate the latent infection in the plant, generating instructions that, when processed by a robotic device, cause the robotic device to (i) navigate to a location of the plant and (ii) apply an antimicrobial treatment to the plant and one or more other plants of similar origin; transmitting the command to the robotic device; The method according to item 1, comprising: (Item 8) determining, by the processor, one or more treatments to mitigate the latent infection in the plant, generating instructions that, when processed by a robotic device, cause the robotic device to (i) navigate to a location of the plant and (ii) harvest one or more plant products from the plant and one or more other plants of similar origin prior to an expected harvest timeframe; transmitting the command to the robotic device; The method according to item 1, comprising: (Item 9) 2. The method according to item 1, wherein the obtained data is generated by a nucleic acid sequencer based on sequencing of plant products extracted from the plant. (Item 10) 10. The method of claim 9, wherein the plant product comprises at least one of a bark, leaf, flower, and fruit sample. (Item 11) Item 12. The method of item 9, wherein the plant product is sampled non-destructively from the plant. 10. The method according to claim 9, wherein the plant products comprise volatile components emitted from the plant. (Item 13) Item 14. The method of item 1, wherein the one or more machine learning models include at least one of a binary logistic regression model, a logistic model tree, a random forest classifier, L2 regularization, partial least squares, and a convolutional neural network (CNN). 2. The method of claim 1, wherein the plant does not contain visible signs of infection. (Item 15) encoding, by the processor, the acquired data into a data structure for input to the one or more machine learning models; providing, by the processor, the encoded data structure as an input to the one or more machine learning models; Item 1. The method according to item 1, further comprising: (Item 16) 10. The method of claim 1, further comprising carrying out, by the processor, one or more of the determined treatments to mitigate the latent infection in the plant. (Item 17) 2. The method of claim 1, wherein the processor's selection of one or more machine learning models is further based on one or more predictive features identified from the acquired data, wherein the one or more predictive features include at least one of a growing region of the plant, environmental conditions, a plant type, a growth stage of the plant, non-invasive measurements of the plant, invasive measurements of the plant, dry matter content, gene expression, emitted volatile components of the plant, a growing zone, and read sequence data of the plant. (Item 18) 2. The method of claim 1, wherein the known plant information includes at least one of growth history information, length of growing season, soil condition, precipitation level, amount of sunlight, environmental temperature, growing region, and plant type for the one or more other plants. (Item 19) A system for predicting latent infection in plants, comprising: one or more processors; one or more computer-readable storage devices storing instructions that, when executed by the one or more processors, cause the one or more processors to perform operations; and the operation comprises: obtaining data describing the expression levels of one or more infection biomarkers present in the plant, wherein the infection biomarkers indicate a likelihood of infection in the plant, and the data indicates a difference between a read sequence of the plant and a reference genome of a healthy plant of the same species as the plant; selecting one or more machine learning models based on the obtained data, the one or more machine learning models having been previously trained using data correlating other data with the identified one or more infection biomarkers of one or more other plants, and generating an output indicative of the likelihood of the plant developing a latent infection before harvest, the one or more machine learning models being trained using a process comprising: inputting into the one or more machine learning models from a training dataset: (i) non-invasive measurements of the one or more other plants; (ii) invasive measurements of the one or more other plants; (iii) known plant information of the one or more other plants; and (iv) positive infection identification information of the one or more other plants; determining a predicted likelihood that the one or more other plants will develop the latent infection before harvest based on inputting (i)-(iv) into the one or more machine learning models; and outputting the one or more machine learning models for runtime use. selecting, generating an output indicative of the likelihood of the plant developing the latent infection before harvest based on applying the one or more machine learning models to the data; determining that the plant has a latent infection based on the output exceeding a predetermined threshold range; determining one or more treatments that reduce the latent infection in the plant; outputting an indication that the plant has the latent infection and the one or more determined treatments; Including, the system. (Item 20) Determining one or more treatments that will reduce the latent infection in the plant includes determining the plant and and one or more other plants of similar origin, identifying at least one of: (i) a modified harvest schedule to earlier in the growing season; (ii) instructions to apply a predetermined amount of pesticide before harvest; and (iii) when the plant and the one or more plants of similar origin are approaching the end of their respective healthy plant productive lives, wherein the one or more machine learning models are pre-trained to identify (i)-(iii) using data from the training dataset.

Claims

[Claim 1] The invention described in the specification.

Images