Plant Variety Selection Using Prediction Models for Extreme Weather
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional methods for selecting plant varieties for cultivation in a target area are time-consuming, expensive, and prone to inaccuracies, especially when dealing with extreme environmental conditions that are not fully observed in field trials, and current genomic selection methods do not account for unobserved environmental factors.
Innovation Solution
A method and system that use a selection score function to predict phenotypic traits based on estimated environmental parameters and phenotype information, allowing for the selection of plant varieties that can perform well under untested conditions, including extreme weather events, without requiring exact environmental data from the target area.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional field testing methods are used to select plant varieties, then variety performance can be observed under actual growing conditions, but the method is time-consuming, expensive, and cannot capture extreme environmental conditions that occur rarely
Solution Approach 1:
The patent applies preliminary action by using prediction models to estimate variety performance under extreme environmental conditions before actual field trials occur. The system uses historical weather data, genomic information, and phenotypic data to predict how varieties will perform under future extreme conditions, allowing breeders to select superior varieties without waiting for rare extreme events to occur during traditional field testing periods
Solution Approach 2:
The patent uses copying by creating virtual representations of extreme environmental conditions through prediction models. Instead of physically waiting for or recreating extreme weather events in field trials, the system copies the effects of extreme conditions through computational models that simulate variety performance under those conditions based on genomic and phenotypic data, thereby capturing rare event impacts without the time and resource costs of actual field testing under extreme conditions
2Measurement precision
If field trials are conducted to observe variety performance, then actual growing conditions are captured, but extreme conditions are rarely observed during the finite testing period
Solution Approach 1:
The patent applies another dimension by adding a computational prediction dimension to the traditional field trial approach. The system uses prediction models that incorporate genomic information, phenotypic data, and historical weather data to estimate variety performance under extreme conditions that fall outside the range of observed field trial conditions. This additional dimensional approach allows the system to assess variety performance across a broader spectrum of environmental conditions, including rare extreme events, without being limited by the finite duration and observed conditions of physical field trials
3Productivity
If genomic selection methods are used to determine superior genotypes, then the breeding cycle is accelerated, but environmental factors affecting plant growth in unobserved growing seasons are not accounted for
Solution Approach 1:
The patent applies merging by combining genomic selection methods with environmental prediction modeling. The system integrates genomic information, phenotypic data, and predicted environmental conditions into a unified selection framework. This merged approach allows the system to maintain the speed advantages of genomic selection while simultaneously accounting for environmental factors that will affect variety performance in future growing seasons, including extreme conditions that have not been observed during the breeding process
Data Source
AI summary
A method of selecting a plant variety for cultivation in a target area includes selecting a selection score function; estimating values of a first set of environmental parameters for a predefined future period of time for the target area and receiving a set of phenotype information including phenotypic trait measurements for a first sub-set of a plurality of plant varieties and a set o environmental parameters for said first sub-set. Furthermore, the method includes determining a prediction model for the phenotypic traits; using the prediction model to output predictions for phenotypic traits for the plurality of chosen plant varieties; using the selection score function to compute selection score values; and selecting at least one plant variety to be cultivated in the target area from the plurality of chosen plant varieties, based the computed selection score values of the plurality of chosen plant varieties.