3 ms·
Crispr tweak may help gene-edited crops bypass biosafety regulation
- nonbel 11y agoLink to actual paper: http://www.nature.com/nbt/journal/vaop/ncurrent/full/nbt.3389.html http://www.nature.com/nbt/journal/vaop/ncurrent/full/nbt.338... See figure 2D of the previous paper from the same group to see the treatment is toxic (at least to some cell types): http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4032847/ http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4032847/ They don't account for that possibility here. They report in the supplements that ~1/100,000 cells are already mutants in the control group (Fig S1). This may be an underestimate since they only detected ~7% mutant cells after treatment using sequencing, but in fig 1c of the paper they report ~70% using the T7 endonuclease assay. I don't see where this discrepancy is discussed, but clearly one of these estimates is not reliable. Anyway, as they write in the abstract: "The targeted sites contained germline-transmissible small insertions or deletions that are indistinguishable from naturally occurring genetic variation." Yes, and they may very well be detecting natural variation by selecting for the pre-existing mutants. Even using the .01 % pre-existing mutant value there would be 10 mutants in the dish of ~10^5 cells. If the treatment kills 99.9% of the normals you would be left with around 10 mutant and 100 normal colonies. This is about the size of the observed effect according to the sequencing method. They need to report how many cells were remaining after the CRISPR-Cas9 treatment and figure out why the T7 endonuclease assay is disagreeing with the sequencing by an order of magnitude.