[nonbel]

I have never seen a paper on CRISPR that can distinguish between selecting pre-existing mutants and actually modifying genes. I have read probably a dozen or so at this point, and it is amazing that they always fail to address this either in citations or actual data.

At first I thought it was an honest mistake, but now it would not surprise me if some of the main players know that their experiments with CRISPR have been misinterpreted. They are then pushing the gene "modification" label anyway because it is sexier.

After all, CRISPR has received an extremely unusual amount of media coverage over the last year or so, which raises red flags. I suspect a marketing effort is being directly funded. That is not a honest use of funds meant for research, especially that which is not meeting minimum scientific standards (ruling out other explanations for the results rather than just a null hypothesis).

[daughart]

It's not clear to me what you are saying. Cas9/CRISPR unambiguously cuts the genome at a target site determined by the guide RNA sequence. In the presence of a DNA oligo with partial complementarity to the cut site, the DNA repair mechanism will sometimes incorporate the "payload" oligo, causing the genetic locus to be engineered from one sequence to another, predetermined sequence.

[nonbel]

Start with 10^6 cells. Say 0.1% (1 in 1000) are already mutants at that site. Then add something that kills 100% the non-mutants and you will be left with 10^3 mutants without any gene editing. Say it kills 50% of the non-mutants and renders the rest quiescent due to DNA damage (not dividing), then you are left with 10^3 mutants and 5 x 10^5 non-mutants at time t0. After eg 7 divisions you will have 10^3 x 2^7 = 1.28 x 10^5 mutants, corresponding to 25% of the total.

It depends on the initial number of cells, initial proportion of mutants, division rates, and toxicity. I have also noted that the initial number of cells is usually reported without any uncertainty, which makes me think those numbers may be rather unreliable.

[daughart]

No pre-existing cells have the mutation at the site you're trying to engineer. It just doesn't happen. Otherwise selection alone would be good enough. But mice cells don't have that much intrinsic variation. Plus a lot of time they're inserting whole genes or larger payloads. The statistical probability of that arising from chance is zero.

[nonbel]

>"No pre-existing cells have the mutation at the site you're trying to engineer. It just doesn't happen."

Not in any paper on CRISPR I have read, in fact just the opposite: there are always low levels of mutants found in the controls (eg Schumann et al 2015 linked below). Please link to the papers that have lead you to make this claim.

[daughart]

That is far more easily explained by contamination, which, as you mention, is actually how they explain it in papers.

[nonbel]

Here is another (supplementary table 2). https://www.ncbi.nlm.nih.gov/pubmed/26121415

I can keep going, but would prefer you bring references of your own so I cannot be accused of cherry picking.

[speeder]

My sister was sent by Brazil's government to MIT so she can bring back CRISPR technology to Brazil public universities to speed a research here about using gene editing to control stem cell expression.

I've seen plenty of people and papers where dna was edited in across entirely different organism realms, bacteria dna in animals, animal dna in plants, and so on...

I honestly don't understand how the technique ins't about editing, the only use I saw for it is editing.

EDIT: I asked my sister to give me links to some papers, I will post them after she replies.

[nonbel]

>"EDIT: I asked my sister to give me links to some papers, I will post them after she replies."

Thanks, I appreciate it. This has been bugging me for awhile now.

[nonbel]

You'll have to link to the exact evidence and methods you are referring to.

[daemonk]

But there are plenty of knock-in experiments where foreign DNA was put into the cut site. Are you saying a pre-existing mutant with the foreign DNA was used in those cases too?

[nonbel]

Please link to one/some of the papers you are referring to. From what I have seen, they always detect edits in the controls or fail to report enough information to say either way, eg:

"Although rare (∼1–2%), edits were detected with Cas9-only control treatment, including at the predicted CXCR4 cut site, potentially indicating trace amounts of experimental contamination of the Cas9 RNPs." http://www.pnas.org/content/112/33/10437.full

Note the curiously missing rows in dataset S1.

[cowsandmilk]

What other controls do you expect to see? Dataset S1 seems complete to me. There is a similar background level of indels both at the CXCR4 site and off-target 1 and off-target 2 sites. The experiment increases indels at the CXCR4 site, but not at the off-target sites.

[nonbel]

The proportion of HDR (ie HindIII reads only) without the template. What percent of cells will randomly mutate to get HindIII recognition sites?

Also, seeing what happens using template only would be good. We would expect low baseline levels of HDR to occur right? I it is plausible few out of 10^5 or 10^6 cells will require repair at that locus even without any Cas9.

[jimrandomh]

The papers I've looked at sequence and measure the on-target mutation rate, and don't have any steps in them that would select for mutants (because that would ruin the measurement). Where do you propose the selection for mutants would be happening?

Unless I'm misunderstanding something about how the experiments are done, your theory would require many groups to be independently committing scientific fraud, which is very implausible.

[nonbel]

No fraud is necessary, just sloppy interpretation of data.

Staying with Schuman et al (2015) linked in this thread, they start with 2.5 x 10^5 cells and end up with 5 x 10^4 to 2 x 10^5 three to four days later. Why are there fewer cells even without accounting for any division? Because the treatment is toxic. This is reported in many papers.

I don't know what the proliferation rate is like for the cells in the conditions of that study, but apparently up to 7 divisions in 4 days is considered plausible for T-cells: http://www.ncbi.nlm.nih.gov/pubmed/17367338

[jimrandomh]

If you're right and CRISPR doesn't actually work, then none of the papers published so far are replicable and you'll be vindicated in mere months. If that doesn't happen, then there's a flaw in the style of reasoning and research that led you to conclude the existing papers were flawed. I recommend jotting down a few notes about how you came to this conclusion, and making a calendar reminder to check how it turned out next year.

[nonbel]

>"If you're right and CRISPR doesn't actually work, then none of the papers published so far are replicable and you'll be vindicated in mere months."

Not at all. I'm not sure you understand what I am saying.

It appears to me the data can be interpreted in multiple ways. Two different "theories" can explain the same results. This is a much more insidious problem than mere non-replication (I haven't seen any direct replications regarding CRISPR either though). People can continue on the wrong path for a long time by interpreting good, reliable data incorrectly.

https://en.wikipedia.org/wiki/Experimentum_crucis

[tstactplsignore]

Lol, what, just what... the biophysics of the Cas9 system are understood pretty well. We literally get as close as you can get to observing the accepted mechanism of action. How would you explain the consistency of site specific gene integration? You can't be selecting from existing mutants and "just happen" to get your transgenic product inserted at the exact site you specified. Your comment is just absurd. You also don't seem to realize that CRISPR is used at pretty much every university in dozens of different systems by thousands of different scientists. What a bizarre comment. Edit: uh, if you actually need a "source" for this:

http://m.pnas.org/content/109/39/E2579.full

[nonbel]

>"You can't be selecting from existing mutants and "just happen" to get your transgenic product inserted at the exact site you specified. Your comment is just absurd."

This is very easy. If you take a very many cells, some small percent will be mutants at any given site (unless you claim zero background rates of mutation, which is absurd and also directly contradicted by the data in these same papers). If you give a treatment that raises/causes the affinity of DNA damaging substances for a certain site, this will selectively damage the DNA of the non-mutant cells. The proliferation of the non-mutants will be suppressed and many will die off. The remaining mutants will proliferate to fill the gap. See my other post for a (very simple) mathematical model of this phenomenon.

This is not absurd at all. It is basic logic and algebra. I will check that paper and get back to you. I actually have not read any using bacterial cells yet, thanks. Also, as far as I know there are no mathematical models of the standard proposed CRISPR mechanism that have been published, if you know of one that would be great.

[tstactplsignore]

That might be possible if CRISPR was just a site specific knock out system.

It's not. Your theory does not explain how site specific gene integration of transgenic products is possible if CRISPR/Cas is not an efficient site specific nuclease. If Cas9 is not cutting the DNA at the specific site so the transgenic product can integrate there, we wouldn't be getting the results seen.

[nonbel]

What experiment in that paper do you think addresses the issue of selection vs modification? Both require the cleavage of specific DNA sequences, that is all I see reported in Gasiunas et al 2012.

[tstactplsignore]

Here, listen. The following two papers conclusively "disprove" your idea. Both use single embryo injection and show multiple successful site specific mutagenesis in groups of no more than 5 to 25 cells.

http://www.sciencedirect.com/science/article/pii/S0092867413...

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3686313/#SD1

[nonbel]

Yang et al (2013): "To assess whether a marker transgene could be inserted into an endogenous locus, we coinjected Cas9 mRNA, sgRNA, and a double-stranded donor vector that was designed to fuse a p2AmCherry reporter with the last codon of the Nanog gene (Figure 2A). A circular donor vector was used to minimize random integrations. To assess toxicity and to optimize the concentration of donor DNA, we microinjected different amounts of Nanog-2A-mCherry vector. Injection with a high concentration of donor DNA (500 ng/ml) yielded mCherry-positive embryos with high efficiency, with most blastocysts being retarded, whereas injection with a lower donor DNA concentration (10 ng/ml) yielded mostly healthy blastocysts, most of which were mCherry-negative. When 200 ng/ml donor DNA was used, 75% (936/1,262) of the injected zygotes developed to blastocysts, 9% (86/936) of which were mCherry-positive (Figure 2C; Table S1)."

So efficiency is inversely proportional to toxicity, they treated many more than 5-25 cells (the selection would occur at the level of the embryo), and there was only 1-10% rate of mutation detection. Also, they used "Superovulated female B6D2F1 mice". This procedure leads to chromosomal abnormalities and probably genetic instability so we would expect elevated presence of mutations at any given site: http://jhered.oxfordjournals.org/content/77/1/39.full.pdf

I'll have to look closer at the HDR aspect though (primers used, etc). But what may be going on that usually they detect the insertion via PCR: there is one primer to a sequence unique to the cassette and another upstream or downstream that should only be in the cells. Then the segment spanning the junction is amplified which supposedly is conclusive evidence of insertion at the correct location. The problem is you can get single primer amplification and also the homology arms required for HDR are likely to contain similar sequences to the "cell-only" primer. Eg: http://link.springer.com/protocol/10.1385%2F0-89603-258-2%3A...

Hwang et al 2013: "On the next day, injected embryos were inspected under stereoscope and were classified as dead, deformed or normal phenotypes. Only embryos that developed normally were assayed for target site mutations"

They don't seem to tell us how many embryos were injected. And that study does not appear to use any type of control group at all. AFAICT, that is exactly the type of study that is consistent with a selection effect.

[tstactplsignore]

I'm not an expert in this area, but I still don't understand how you manage to reconcile site specific transgene insertion. In these studies reporter genes are clearly inserted and heritable. How is there anything more to the discussion?

Like, as reported by Doudna:

http://science.sciencemag.org/content/337/6096/816.short

"We show here that in a subset of these systems, the mature crRNA that is base-paired to trans-activating crRNA (tracrRNA) forms a two-RNA structure that directs the CRISPR-associated protein Cas9 to introduce double-stranded (ds) breaks in target DNA. At sites complementary to the crRNA-guide sequence, the Cas9 HNH nuclease domain cleaves the complementary strand, whereas the Cas9 RuvC-like domain cleaves the noncomplementary strand. The dual-tracrRNA:crRNA, when engineered as a single RNA chimera, also directs sequence-specific Cas9 dsDNA cleavage."

Which part of that mechanism do you doubt? It sounds like you doubt the dsDNA nuclease activity of Cas9. Why not just order a plasmid, some Cas9 + gdna, put them together and sanger sequence your products? If Cas9 isn't a site specific guided endonuclease you could prove it for $200.

[nonbel]

>"It sounds like you doubt the dsDNA nuclease activity of Cas9."

Not at all. This would be why the treatment is toxic and suppressive of proliferation.

At this point, I still think the presence of indels at the site (ie the proposed NHEJ mechanism) is just as easily explained by selection for pre-existing mutants. The experiments involving insertion of DNA (ie the proposed HDR mechanism) are better, but lack controls for "off-target" PCR amplification when showing the gels. IE we need to know how often the template itself will be amplified under their primers/conditions, both free and if it gets incorporated in some random location.

When segments across the insertion junction are amplified, sequenced, and reported, I find this convincing as it is a precise prediction that matches the data and I can think of no other explanation. The other experiments are pretty much redundant and add nothing. However, the reports I have seen contain little methodological or quantitative information regarding these sequences which does make me remain skeptical, especially about claims of efficiency. Those claims seem to always be determined using the former experiments that can be explained in other ways.

[Obi_Juan_Kenobi]

This is a baseless critique.

No one has to address alternative hypotheses that don't make any sense. Are you unaware that Sanger Sequencing exists, and the actual lesions can be read? Or that heterologous genes are being introduced with CRISPR methods? Neither of these common results can be explained by the spontaneous insertion of hundreds of nucleotides that happen to precisely match the sequence of the construct being inserted.

Again, this is utter nonsense and demonstrates a complete absence of basic understanding in molecular biology.

[nonbel]

>"Are you unaware that Sanger Sequencing exists, and the actual lesions can be read? Or that heterologous genes are being introduced with CRISPR methods? Neither of these common results can be explained by the spontaneous insertion of hundreds of nucleotides that happen to precisely match the sequence of the construct being inserted."

The first is just as consistent with the selection mechanism, because low levels of baseline mutants ARE reported (see the Schumann et al and Hendel et al papers I linked to in this thread for examples).

The second would indeed be difficult to explain with a selection mechanism, unfortunately I have not seen that actually published. Instead, the primers used can just as well be amplifying the template and/or the sequence of the inserted cassette is not shown (eg Figure 2D and S1 here: http://www.sciencemag.org/content/348/6233/442 )

If you have a reference to a specific paper I would appreciate it.

[daemonk]

What do you mean by the primers can just well be amplifying the template? How does that explain knock-ins without an actual insertion? And there are plenty of knock-in CRISPR papers out there. Just literally search for "CRISPR knock-in".

[nonbel]

Let me ask this. Say you have sequence A that is not supposed to exist before your treatment and sequence B that you have added to the environment in large amounts. Is it safe to use primers where one matches exactly to sequence B and the other is this similar?

CTCATTAGGCACCCCAGGCTTTACA

CTCAGT------CCCAGGCTTTACA

[daemonk]

Are you suggesting that they are just detecting the un-incorporated foreign DNA after CRISPR? I think the fact it has been shown that the knocked-in DNA is inherited to the progeny is strong enough evidence that the DNA was actually inserted.

Unless you want to argue that the un-incorporated DNA was also transmitted to the next generation, which honestly, is extremely unlikely.

[nonbel]

That could possibly explain some results, but not those involving transmission. If the knocked-in DNA is transmitted to the next generation then I'd think it must have gotten incorporated somewhere, however, this need not be at the intended site if the primers are amplifying the template.

Then again, supposedly shingles is caused by extragenomic Varicella-zoster DNA that is somehow stable for decades and can be passed on during pregnancy. I'm not sure I believe that though, and of course that is viral DNA.

Anyway, in that Ruan et al (2015) they claim to have detected exactly the expected sequence across the junction in at least a few cells. I can't think of any explanation for that data other than CRISPR working as advertised. However, they don't report in what percent of the cells this was observed.

Edit: I mean supposedly those exact sequences shown in figure S2 never physically existed before and now they do, exactly as predicted by the theory. That is strong evidence.

[nonbel]

Ok, I did that search and here is the first paper I found: http://www.nature.com/articles/srep14253

EDIT: Let me look again at this paper later.

[nonbel]

Here is another. At first you may think they show successful insertion of GFP, but it turns out no! Instead all that needed to happen was mutation of an early stop codon: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3712628/

That is why I am asking others to provide their own references. The papers I have read do not seem capable of distinguishing between modification vs selection after careful inspection.