Testing for the unintended or adventitious presence (AP) of genetically modified (GM) events in seed lots is a common practice to comply with regulatory requirements and good stewardship practices. A subsample of a seed lot is typically tested for AP levels, and then statistical methods are used to estimate the upper level of AP in the remainder of the lot with a given level of confidence. For large seed lots, a binomial distribution is typically assumed, but for seed lots where the tested sample is a substantial proportion of the overall seed lot, a hypergeometric distribution is typically assumed. Due to the destructive nature of AP seed testing, we suggest that this latter method may overestimate confidence of low AP in the remaining seed.
