Optimized breeding strategies for multiple trait integration: III. Parameters for success in version testing.

Multiple trait integration (MTI) is the process by which a target hybrid (or variety) is converted to add value-added traits to the comprehensive performance package represented by that genotype. The goal is to recover all the attributes of the target hybrid, with the addition of the specified value-added traits. In maize, this process utilizes the backcross breeding method to incorporate transgenic events (or genes) of interest. Thus, MTI involves four main steps: single event introgression, event pyramiding, trait fixation, and version testing to ensure recovery of equivalent performance with at least one version of the converted hybrid. Based on a case study involving conversion of a target hybrid for 15 transgenic events (the female inbred parent was converted for 8 events, and the male inbred parent was converted for 7 events), we explored parameters in version testing to facilitate a high likelihood of recovering at least one version of the hybrid conversion with yield performance equivalency within 3 % of unconverted target hybrid. Using computer simulation, we explored the impact of two factors on the success rate of the MTI outcome: (1) the amount of residual NRP (non-recurrent parent) germplasm remaining in the converted hybrid and (2) the number of versions of each parental line conversion created. A range of residual NRP germplasm from 120 to 180 cM (which represents 95.0-96.6 % germplasm recovery of the target hybrid in maize) and up to 5 versions of each parental conversion were considered, with all possible hybrid combinations of each version of female and male RP (recurrent parent) conversion evaluated for yield. With 5 versions of each RP conversion and testing of 25 hybrid versions, a >95 % rate of success was realized when the amount of residual NRP germplasm in the hybrid conversion was ≤180 cM. When hybrid conversions contained ≤120-cM residual NRP germplasm, only 4 versions of one of the parental conversions were needed (rather than 5), requiring 20 versions of the hybrid conversion to be yield tested. These results have implications in the strategic design of an overall conversion program and for the upstream MTI process, especially in setting thresholds for the amount of NRP germplasm remaining in RP conversions. Furthermore, these results validate findings of (Peng et al. in Mol Breed 33:189-104, 2014a. doi:10.1007/s11032-013-9936-7, in Mole Breed 33:105-115, 2014b. doi:10.1007/s11032-013-9937-6) which outline effective breeding strategies to optimize earlier steps in MTI (preceding version testing).