Since maize is one of the most important cereal crops in the world, establishment of an efficient genetic transformation system is critical for its improvement. In the current study, several elite corn lines were tested for suitability of Agrobacterium tumefaciens-mediated transformation by using immature embryos as explants. Infection ability and efficiency of transformation of A. tumefaciens sp. strains EHA105 and LBA4404, different heat treatment times of immature embryos before infection, influence of L-cysteine addition in co-cultivation medium after transformation, and how different ways of selection and cultivation influence the efficiency of transformation were compared. Glyphosate-resistant gene 2mG2-EPSPS was transformed into several typical maize genotypes including 78599, Zong 31 and BA, under the optimum conditions. Results showed that the hypervirulent Agrobacterium tumefaciens sp. strain EHA105 was more infectious than LBA4404. Inclusion of L-cysteine (100 mg L-1) in co-cultivation medium, and heating of the immature embryos for 3 min prior to infection led to a significant increase in the transformation efficiency. Growth in resting medium for 4-10 d and delaying selection was beneficial to the survival of resistant calli. During induction of germination, adding a high concentration of 6-BA (5 mg L-1) and a low concentration of 2,4-D (0.2 mg L-1) to regeneration medium significantly enhanced germination percentage. Using the optimized transformation procedure, more than 800 transgenic plants were obtained from 78599, Zong 31 and BA. By spraying herbicide glyphosate on leaves of transgenic lines, we identified 66 primary glyphosate-resistant plants. The transformation efficiency was 8.2%. PCR and Southern-blot analyses confirmed the integration of the transgenes in the maize genome.

In this paper, the many indices used in validation of crop models, such as RMSE (root mean square errors), Sd (standard error of absolute difference), da (mean absolute difference), dap (ratio of da to the mean observation), r (correlation), and R2 (determination coefficient), are compared for the same rice architectural parameter model, and their advantages and disadvantages are analyzed. A new index for validation of crop models, dap between the observed and the simulated values, is proposed, with dap<5% as the suggested standard for precision of crop models. The different kinds of validation methods in crop models should be combined in the following aspects: (1) calculating da and dap; (2) calculating the RMSE or Sd; (3) calculating r and R2, at the same time, plotting 1:1 diagram.