After reading the paper[1], AFAICT the phenotype results can also be explained by selecting for pre-existing mutants (the cells of the embryo containing the targeted sequence are killed off) followed by transvection[2]. They assume a simple dominant-recessive relationship between alleles of the yellow gene and pigmentation, which is not true[3]. Also, the primers they used for PCR shown in Fig 2D don't look like they span the genome-insertion junction...<p>[1] <a href="http://www.sciencemag.org/content/348/6233/442" rel="nofollow">http://www.sciencemag.org/content/348/6233/442</a><p>[2] <a href="https://en.wikipedia.org/wiki/Transvection_%28genetics%29" rel="nofollow">https://en.wikipedia.org/wiki/Transvection_%28genetics%29</a><p>[3] <a href="http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2766311/" rel="nofollow">http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2766311/</a>