Li Zhang1, Melanie Correll2, Carlos Vallejos3, Kenneth J. Boote4, Senthold Asseng2, James W. Jones5, Wei Hou6, Jose Alejandro Clavijo Michelangeli7 and Mehul Bhakta7, (1)University of Texas-Austin, Austin, TX (2)Agr. & Biol. Engineering Dept., University of Florida, Gainesville, FL (3)Horticultural Sciences, University of Florida, Gainesville, FL (4)Agronomy Dept., 3105 McCarty Hall, University of Florida, Gainesville, FL (5)Ag. and Bio. Engineering, University of Florida, Gainesville, FL (6)Preventive Medicine, Stony Brook University Medical Center, Stony Brook, NY (7)Agronomy, University of Florida, Gainesville, FL
Models that can accurately predict bean growth and development, based on the genetic makeup of the cultivar, can be useful to breeders in the design of cultivars adapted to specific environments. Our goal is to identify quantitative trait loci (QTL) controlling vegetative phase traits of the common bean (Phaseolus vulgaris) under a range of environments, and to incorporate this genetic information into a model that predicts the vegetative growth ( a gene-based model). To achieve this goal, we have characterized vegetative phase traits that are segregating in a recombinant inbred family. This family was grown in North Dakota, Florida, Puerto Rico, and two sites in Colombia. We have tested a model that assumes node addition during the vegetative phase is only affected by temperature – dN/dt = Nm × f(T), where dN/dt is the rate of node addition per day, Nm is the maximum node/leaf appearance rate under optimum condition, and f(T) is the cardinal temperature function; dN/dt for different sites should be similar if calendar days are adjusted using thermal time or physiological day. Following this model, we estimated the base temperature (Tbase), first optimum temperature (Topt1), and Nm (maximum rate of node addition/leaf appearance rate) for each of 97 RILs during their vegetative phase. The average Tbase and Topt1 for these RILs were 8.8°C and 22.8°C, respectively. Meanwhile, difference of Nm for determinate and indeterminate lines in the family was found (0.25, 0.32 node/physiological day on average for determinates and indeterminates, respectively), and one QTL associated with the fin gene, which controls growth habit in bean, was detected for Nm. Thus, the node addition model can be improved by incorporating the FIN gene into the parameter Nm. Additional early vegetative traits and their associated genes are being identified to incorporate them into other relevant models of early plant development.