August 13th, 2020

1. Introduction to the problem

Crop by region

ICRISAT-Sorghum Asia

Problem specification

Line breeding in Sorghum is not exploiting the full potential of non-additive variation.

Breeding strategy component tackled

All

Breeders’ equation terms tackled

\(\Delta_g = (i * \sigma_g * r)/L\)

Hypothesis

Introducing effective hybrid breeding will deliver materials with higher performance.

2. Materials and methods

Treatments

Treatment Description
PERSE_PYT_AYT Current line breeding scheme, selecting parents at PYT (50), AYT (20) using a base index.
TESTER1_PYT_AYT Hybrid breeding scheme, selecting parents at PYT (50; using GCA from 1 tester), AYT (20; using GCA from 5 testers) using a base index.
TESTER2_PYT_AYT Hybrid breeding scheme, selecting parents at PYT (50; using GCA from 2 testers), AYT (20; using GCA from 5 testers) using a base index.
TESTER5_PYT_AYT Hybrid breeding scheme, selecting parents at PYT (50; using GCA from 5 testers), AYT (20; using GCA from 5 testers) using a base index.

Simulation procedure

A 20 year burn-in period was modeled using the line breeding schemes. The burn-in was followed by a 50 year evaluation period to measure rates of genetic gain for all treatments. Genetic gain was measured by assessing changes in genetic merit in PYT hybrids among pools. Genotype-by-year interaction variance was assumed to be equivalent to genetic variance (based on average correlation between locations being equal to 0.5). 20 replications done. We simulated 5 complex, 5 intermediate and 5 simple traits to be behind the genetic merit.

2. Genetic model of hybrid breeding

3.1 Results: Is there an increase in hybrids among pools vs within pools? (meanDD=0.2)

Based on theory we expect some difference as time evolves. By year 50, the difference of within-pool vs among-pool crosses is 1.07 (95% CI: 1.06,1.07) times more gain.

3.2 Results: Is there advanatage of hybrid breeding over line breeding? (meanDD=0.2)

Based on theory we expect that dominance will imply that independent of the selection strategy we will observe hybrids being better than lines. Assuming a partial dominance by year 50, the difference of lines vs among-pool crosses is 1.45 (95% CI: 1.44,1.47) times more gain.

3.3 Results: which selection method produces higher performance hybrids? (meanDD=0.2)

Based on theory we expect that dominance will allow GCA selection strategies to be better. By year 50, the difference of among-pool crosses for different selection methods yielded. GCA based on 2 testers has 1.01 (95% CI: 1.01,1.02) times more gain than selection based on per-se performance.GCA based on 5 testers has 1.01 (95% CI: 1.01,1.02) times more gain than selection based on per-se performance. GCA based on 5 testers has 0.98 (95% CI: 0.98,0.99) times more gain than selection based on GCA with 1 tester.

4. Conclusion

Given the existence of dominance in the gene action of Sorghum populations and the possibility to create seed hybrid breeding is a good option.

We recommend moving completely to hybrid breeding based on either reciprocal recurrent selection scheme (using a GCA/SCA approach), or independent pools selected by per-se performance depending on the level of dominance.