Prediction of infection in plant products

A system for identifying a latent infection in a plant product comprising:

• at least one processor;
• and
• a memory device storing instructions that are operable, when executed by the at least one processor one or more computers, to cause the at least one processor to perform operations comprising:
• obtaining data describing a level of expression of one or more infection biomarkers of a plant product, the infection biomarkers indicating a likelihood of infection in the plant product, wherein the data identifies one or more variants that describe differences between a read sequence of the plant product and a reference genome of a healthy plant product;
• encoding the obtained data into a data structure for input to a machine learning model;
• providing encoded data structure as input to the machine learning model, wherein the machine learning model was previously trained, using data that correlates other encoded data structures and determined one or more infection biomarkers of one or more other plant products, to predict likelihoods of infection in the one or more other plant products, wherein the machine learning model was trained using a process comprising:
• inputting, from a training data set, (i) a first set of training samples of infection biomarkers of the one or more other plant products and (ii) known rates of infection for the first set of training samples into the machine learning model,
• determining the predicted likelihoods of infection in the one or more other plant products based on inputting (i) and (ii) into the machine learning model,
• determining a difference between the predicted likelihoods of infection and the known rates of infection,
• generating, based on the determined difference, a set of parameters for validating the machine learning model,
• validating, using the set of parameters, the machine learning model so that the determined difference between the predicted likelihoods of infection and the known rates of infection are within a predetermined threshold range, wherein validating the machine learning model comprises inputting a second set of training samples from the training data set into the machine learning model, and
• outputting the validated machine learning model for runtime use;
• obtaining, from the machine learning model, the generated output data indicating a likelihood that the plant product has a latent infection;
• determining, based on the generated output data, that the plant product has a latent infection;
• and
• performing one or more operations to mitigate the latent infection in the plant product.

A non-transitory computer-readable medium storing instructions stored thereon that, when executed by at least one processor of a computing device, cause the at least one processor of the computing device to perform operations comprising:

• obtaining data describing a level of expression of one or more infection biomarkers of a plant product, the infection biomarkers indicating a likelihood of infection in the plant product, wherein the data identifies one or more variants that describe differences between a read sequence of the plant product and a reference genome of a healthy plant product;
• encoding the obtained data into a data structure for input to a machine learning model;
• providing encoded data structure as input to the machine learning model, wherein the machine learning model was previously trained, using data that correlates other encoded data structures and determined one or more infection biomarkers of one or more other plant products, to predict likelihoods of infection in the one or more other plant products, wherein the machine learning model was trained using a process comprising:
• inputting, from a training data set, (i) a first set of training samples of infection biomarkers of the one or more other plant products and (ii) known rates of infection for the first set of training samples into the machine learning model,
• determining the predicted likelihoods of infection in the one or more other plant products based on inputting (i) and (ii) into the machine learning model,
• determining a difference between the predicted likelihoods of infection and the known rates of infection,
• generating, based on the determined difference, a set of parameters for validating the machine learning model,
• validating, using the set of parameters, the machine learning model so that the determined difference between the predicted likelihoods of infection and the known rates of infection are within a predetermined threshold range, wherein validating the machine learning model comprises inputting a second set of training samples from the training data set into the machine learning model, and
• outputting the validated machine learning model for runtime use;
• obtaining, from the machine learning model, the generated output data indicating a likelihood that the plant product has a latent infection;
• determining, by the processor and based on the generated output data, that the plant product has a latent infection;
• and
• performing, by the processor, one or more operations to mitigate the latent infection in the plant product.