Dynamic crop simulation models can be transformed into gene-based models by replacing an existing process module with a gene-based module for simulating the same process.
Dynamic crop simulation models are tools that predict the phenotype (i.e. observable characteristics) of plants grown in specific environments. In these models, genotypic differences among cultivars are represented by empirical genotype-specific parameters (GSPs).
“Traditionally, genotypic information for each cultivar has to be estimated from experimental data that are collected from multiple field studies that represent different environments. Advances in technologies for rapidly and inexpensively identifying genetic makeup of plants have now made it possible to integrate information on variations in genes among the cultivars into these models,” says Dr. Gerrit Hoogenboom from the University of Florida
Hoogenboom led a study recently published in in silico Plants that incorporated a dynamic gene-based module into an existing crop model to create a “hybrid dynamic model.”
“The current challenges are that for each biophysical process we need to estimate the dynamic model parameters as a function of QTL and environmental data. However, once this has been done, we only need the QTL information for new cultivars or hybrids to simulate growth and development and predict yield, compared to the traditional approach that requires extensive experimental data collection prior to any model application with new cultivars,” says Hoogenboom.
The gene-based flowering module was calibrated using 12 QTL (quantitative trait loci, which are DNA regions) that were previously found to control the time to flowering to create a gene-based flowering module. The original flowering component in the CSM-CROPGRO-Drybean model was then replaced by the gene-based module. Simulations with QTL and weather as inputs for the module were conducted for multiple locations to predict developmental timing using the QTL-based module and the CSM-CROPGRO-Drybean model was used for predicting the other processes and ultimately yield.
The integrated gene-based hybrid module simulated days of first flower that agreed closely with observed values. The hybrid model also described most of the gene, environment, and gene x environment interaction effects on time-to-flower and was able to predict final yield and other outputs simulated by the original model.
The approach used for integrating the gene-based first flower module into the CSM-CROPGRO-Drybean model can potentially be used to incorporate other gene-based modules to systematically transition from a GSP- to a gene-based model.
See the original post