1 Modelling spatial distribution of outcrossing rates between neighboring maize fields

Spatial distribution of outcrossing rates between neighboring maize fields was modeled in order to determine the most important factors influencing the gene flow. The machine learning technique of regression tree induction was used to build models that predict the degree of outcrossing from data on the relative position of the donor and recipient fields and the winds, as well as several phenological variables. The resulting models show that the distance between the fields plays a dominant role, followed by the angle and the percentage of appropriate wind (blowing from the donor to the recipient field). Key words Genetically modified crops, gene flow, cross-pollination, maize, machine learning, regression tree