October 8th, 2020
The scope of the simulations to come: accuracy
The scope of the simulations to come: intensity
The scope of the simulations to come: cycle time (skipping stages in PD)
The scope of the simulations to come: cycle time (early recycling)
1. Introduction to the problem
Crop by region
CIAT-Beans EA
Problem specification
Maximum accuracy is not being exploted to increase rates of genetic gain.
Breeding strategy component tackled
Crossing, Evaluation, Selection
Breeders’ equation terms tackled
i
\(\Delta_g = (i * \sigma_g * r)/L\)
Hypothesis
Using GBLUP in the selection process by increasing the intensity at recycling stages may increase the rate genetic gain.
2. Materials and methods
Treatments
| Treatment | Description |
|---|---|
| P:1 - G:1 | Phenotype 200, Genotype 200, Selection population 20/200. Selecting parents at PYT_F5 (20/200), VEF_F8 (5/60), OF_F9 (10/24) using a base index. |
| P:1 - G:2 | Phenotype 200, Genotype 400, Selection population 20/400. Selecting parents at PYT_F5 (20/200), VEF_F8 (5/60), OF_F9 (10/24) using a base index. |
| P:1 - G:2.5 | Phenotype 200, Genotype 500, Selection population 20/500. Selecting parents at PYT_F5 (20/200), VEF_F8 (5/60), OF_F9 (10/24) using a base index. |
| P:0.75 - G:1 | Phenotype 150, Genotype 200, Selection population 20/200. Selecting parents at PYT_F5 (20/200), VEF_F8 (5/60), OF_F9 (10/24) using a base index. |
| P:0.5 - G:1 | Phenotype 100, Genotype 200, Selection population 20/200. Selecting parents at PYT_F5 (20/200), VEF_F8 (5/60), OF_F9 (10/24) using a base index. |
Simulation procedure
A 20 year burn-in period was used. Burn-in was followed by a 30 year evaluation period to measure rates of genetic gain in F9 lines. Genotype-by-year, genotype-by-location interaction variances were assumed to be equivalent to main genetic variance. 20 replications done. We simulated 5 complex and 3 simple traits to be behind the genetic merit.
The targeted change
The different models to estimate the surrogates of genetic value are applied at the recycling stages (PYT & VEF).
3. Results in CIAT-Beans (genetic gain)
- Increasing intensity (100% more) will provide 1.04 (95% CI: 1.02,1.06) times more gain.
- Decreasing intensity (25% less) will provide 0.98 (95% CI: 0.97,1) times more gain.
The genetic model (r, R and \(\sigma_g\))
Accuracy: All treatments don’t seem to vary much in accuracy despite programs being smaller (good).
Response to selection: Size also don’t have a big impact in R compared to baseline.
The genetic model (in % increase/decrease)
Accuracy: Small pop size (i.e. 10/cross) at PYT affects the accuracy performance.
Response to selection: Small pop size (i.e. 10/cross) at PYT affects R so may be risky to use GS for intensity at small pop sizes.
4. Conclusion
Using GBLUP to play with the intensity of the program can be good to free some resources and keep the accuracy and response to selection close to the current levels even for multi-trait scenarios.
Number of progeny per cross at PYT is an important consideration to apply this method.
In the CIAT-Beans program using GBLUP to increase intensity at PYT showed that genotyping more individuals to increase intensity can have marginal gains (1.02-1.04 times more gain). But that can be used to decrease phenotyping and fill the gaps with genotyping (0.98 - 0.99 times more gain).