Nat. Biotechnol. 24, 23–25 (2006), corrected 27 March 2006

In the print version of this article and the version published online, several inaccuracies were introduced into the text through editorial modifications. On page 23 introduced...” should have read “...more than five years....” On page 24, column 1, rows 35–37: “Currently, the method of choice for transgenic sequence detection is real-time PCR.” should have read “...for transgenic sequence quantification....” >Page 24, column 1, rows 51–55: “CRMs are prepared from certified seeds of the GM inbred line and of the non-GM parental variety used for the transformation are homogenized and grinded to powder.” should have read, “For the GM inbred line and the non-GM parental variety used for the transformation, CRMs are prepared from certified seeds, homogenized and grinded to powder.” On page 25, column 1, rows 22–25: “The EU legislators continue to fudge; the current regulation 1829/2003 (ref. 4) uses the same imprecise 1% threshold as its predecessor 49/2000 (ref. 2). And the EC's most recent recommendation (2004/787/2000)17 only partly....” should have read “....uses the same imprecise threshold definition as its predecessor 49/2000, which introduced a 1% threshold2. And the EC's most recent recommendation (2004/787/EC) only partly...” The errors have been corrected in the PDF version of the article.