Modeling some mineral nutrient requirements for micropropagated wild apricot shoot cultures

A response surface methodology (RSM) experimental design was applied for improving micropropagation of a wild apricot, Prunus armeniaca Lam., from the mountains of Kazakhstan. In an initial study, woody plant medium (WPM) mineral nutrients calcium nitrate, ammonium nitrate, mesos (calcium chloride, potassium phosphate and magnesium sulfate) potassium sulfate and minor nutrients] were tested in a response surface methodology (RSM) experiment. Shoot quality was the best when nitrogen and mesos (CaCl₂, MgSO₄, K₂SO₄, KH₂PO₄) compounds were altered. In this study an expanded mesos optimization experiment was run. Data taken included a subjective quality rating, shoot length, shoot number, leaf color and size, callus and physiological disorders. Data were analyzed by Classification and Regression Tree Analysis (CART), a data mining technique that provides specific cutoff values for data and easy to interpret data trees. The CART analysis indicated that the best quality would be with ≤2.4× WPM levels of KH₂PO₄ and ≤0.75× MgSO₄. Shoot length was affected by K₂SO₄, but most shoots were of good size at any concentration. Shoot multiplication was affected by KH₂PO₄, but there were >5 shoots at any concentration. Leaf color was best with ≤2.41× KH₂PO₄ and ≤1.22× K₂SO₄. Based on the CART analysis, a recommendation for improved mesos compounds was developed. Each of the individual trees was analyzed and the cutoff points determined so that all the growth characteristics could be considered in the final concentrations chosen. Using the combined results from the CART analysis, the suggested medium would include WPM with CaCl₂ 2.7×, MgSO₄ 2.7×, K₂SO₄ 0.8×, KH₂PO₄ 0.75×.