[vbuwivbiu]

Exactly this. Every thread on GM quickly polarizes into useless for and against sides, when we need to be selective about it.

And I've never bought the characterization of breeding/selection as a form of GM. Breeding is selection of a pathetically tiny set of modifications, but GM is on a whole other level. With GM we can write arbitrary binary strings of code into any organism. That's simply not comparable with breeding -- the space of realizable phenotypes is orders of magnitude larger with GM.

[Laforet]

Modern breeding involves actively mutating known species with DNA-altering chemical mutagen and ionising radiation. Think fuzzing vs. static code analysis, which one would you do first when it comes to debugging?

>Breeding is selection of a pathetically tiny set of modifications, but GM is on a whole other level.

The truth is probably closer to the opposite of this: most GM species in the pipeline are rather unimaginative ones improving one phenotype at a time, and the vast majority of them focus on pest or herbicide resistance. There is a couple of near commercial crops aimed at improving nutritional content such as golden rice and high-lysine corn, but they are more exceptions than the rule.

The kind of totally disruptive GM e.g. introduce C3 photosynthesis in C4 plants, has not left the drawing board yet for a good reason.

[vbuwivbiu]

Your argument is that GM is less powerful than breeding because people have been "unimaginative" with it ?

OK if you wanted to create, say, a fluorescent horse, which technique would you use:

* GM

* breeding with radiation and chemical mutagens

You have one month.

[Laforet]

Neither technique will work in the timeframe you specified because gestation time for horses are almost a year long.

On the other hand, one can slowly breed for desirable traits such as speed, endurance or disease resistance in horses, good luck doing that with GM.

[vbuwivbiu]

The point wasn't how long it would take for the organism to gestate, it was how long it would take for the trait to be introduced into the organism.

We can do that precisely for fluorescence and many other traits with GM today, but good luck trying to breed a fluorescent horse, it could take millions of years if at all.

[Laforet]

There is no reliable or efficient vector to introduce fluoresence to somatic cells in a horse-sized animal, so no, 1 month is still way too short. You seem to have a highly romanticised idea about the capability of GM but it is misguided.

Besides, GFP fluoresence is the proverbial low hanging fruit as it is robustly expressed in many cell types but has very little practical value. Inserting useful traits via GM is often no short cut because they are rarely determined by a single gene.

[vbuwivbiu]

the point isn't about actually growing an adult fluorescent horse, it's about proving the principle in an embryo - which takes days. It's already been done with mice, but I agree it's a simple example.

Your point about multiple genes is a very good one but we have a good understanding of how some gene networks work already and the rest will follow.

[nixonpjoshua1]

I agree that the polarization is useless. However what you said about selective breeding vs GM is just wrong. Selective breeding introduces numerous mutations, including ones often not even related to the desired traits. In contrast human directed genetic modification often changes a single gene, the state of the art in research universities for specific modifications is only in the tens of genes. You can only say they are not comparable in the sense that selective breeding introduces far more changes than GM and also in the sense that with GM we know what changed where in selective breeding we do not necessarily know how the change was produced or what the extent of it was.

Also I understand that this is a computer centered site but thinking of DNA as binary is a often used but terrible analogy. Living organisms were designed through complete randomness, try refactoring that. If you wanted to turn teosinte into corn without any random mutations and selection I think that would be far beyond our abilities for 100 years at least.

[vbuwivbiu]

Sorry but I didn't compare evolution with computer programming - it's very different to human code, but it is code and it is a computation.

In evolution there's no design, which implies prediction and a model of function. But evolution is far from "completely random". In fact evolution goes to great lengths to correct mutations and only allows mutation in certain carefully controlled sequences of the genome where variation is potentially useful. There are sequences in our genomes that are identical to those of yeast, having been faithfully copied, and error-corrected, for billions of years.

[legulere]

> In contrast human directed genetic modification often changes a single gene

Even CRISPR-CAS9 is not perfect and introduces a lot of unwanted genetic changes when used. Further the big advantage of GM compared with selective breeding is that you can inject Transgenes. This makes them totally incomparable.

[vbuwivbiu]

give it a chance - it's early days, and there are many other techniques and more in the works

[colordrops]

yes, it's absurd that they bucket artificial selection and gene transfer into the same category.