02689nam a2200289 a 450000100080000000500110000800800410001910000130006024501000007326001670017330000130034052017900035365000370214365000100218065300280219065300180221870000200223670000140225670000160227070000150228670000160230170000210231770000140233870000140235270000150236670000180238110578732017-12-12       2017    bl uuuu     u01u1 u    #d1 aLADO, B.  aComparing strategies to select crosses using genomic prediction in two wheat breeding programs.  aIn: International Wheat Genetics Symposium, 12, Tulln, Austria; April 23-28, 2017; BOKU: University of Natural Resources and Life Sciences, Vienna, Austria.c2017  ap.88-90.  aKey message: Evaluation of crosses prediction methods with and without accounting for progeny variance. Mid-parent values was a much larger factor determining genetic gain than increasing the progeny variance of a cross.   In wheat breeding programs, a critical decision is to determine crosses that have high probability to deliver progenies with higher genetics gains (Zhong & Jannink 2007, Bernardo 2014). We present an application of genomic models for predicting parental cross combinations for grain yield, grain protein, and loaf volume across two wheat-breeding programs, INIA-Uruguay and CIMMYT. We evaluated three methods for selecting the ?best? crosses based on (1) mid-parents, (2) top 10% of the progeny within a cross, and (3) maximizing mean and variance within progeny using thresholds. The last two methods were evaluated with the predicted variances obtained through progeny simulation using the PopVar (Mohammadi et al. 2015, Tiede et al. 2015) package in R software. The first two methods showed 82% of crosses in common for yield, 55% for loaf volume and 53% for grain protein, even though only the second method accounts for the variance of the progeny (Figure 1). While the expected variance of the progeny is important to increase chances of finding superior individuals from transgressive segregation, we observed that the mid-parent values of the crosses selected was a much larger factor determining genetic gain than increasing the progeny variance of a cross (Figure 2). Overall, the genomic resources and the statistical models are now available to plant breeders to predict both the performance of breeding lines per se as well as the value of progeny from any potential crosses, but further understanding on optimizing the cross combinations is needed.  aMEJORAMIENTO GENETICO DE PLANTAS  aTRIGO  aWHEAT BREEDING PROGRAMS  aWHEAT QUALITY1 aBATTENFIELD, S.1 aSILVA, P.1 aQUINCKE, M.1 aGUZMAN, C.1 aSINGH, R.P.1 aDREISIGACKER, S.1 aPEÑA, J.1 aFRITZ, A.1 aPOLAND, J.1 aGUTIERREZ, L.