Elucidate CRISPR to a physics-fag:

I study physics and genetics obviously isn't my forté, so could someone with knowledge of genetics explain CRISPR to me, along with its pros and cons?

Other urls found in this thread:

stuffyoushouldknow.com/podcasts/how-gene-editing-works/
twitter.com/SFWRedditGifs

stuffyoushouldknow.com/podcasts/how-gene-editing-works/

A little popsci-ish, but not a bad overview.

Thank you.

How do you do this for every single cell?

let me guess
>Kurzgesagt

You don't. You do it in one cell and then propagate that cell.

In cell culture, you can plate out cells on a selective media so that only cells with the plasmid containing Cas9 and/or your guide RNA will survive. If you're doing it in something like worms or embryos or whatever, you can just shove in the gRNA and as long as Cas9 is present it'll make edits. Small RNAs are very mobile in worms, and in embryos you can just catch them at an early stage so you get a chimeric animal that hopefully includes germline cells.

In plants, you use Agrobacterium to infect either ovules in developing flowers or induced callus (basically plant stem cells) and then either wait for seed or regenerate the callus into a new plant.

Ingenious; why is everyone throwing a strop about it?

because it lets you make highly specific double-stranded breaks with ease. Lots of people are using this to knock out genes - you create a double-stranded break even in an early portion of your gene, some of the time that's repaired by non-homologous end-joining (which is pretty sloppy most of the time) and you end with and insertion or deletion of a few bases. If the size of the indel isn't a multiple of 3, you've fucked up the entire coding frame for the gene and you won't make any functional product.

there's lots of other neat shit you can do but that's the most straightforward one.

and like i said, this is piss-easy. you can train an undergrad how to design highly specific guides in a couple days, and the cloning uses standard techniques. hell, now there's databases where people have already gone through and found all the unique or highly specific regions in the genome of any organism you please that you can target.

we've had tech like this before but it was shitty and never worked. zinc fingers, TALENs, whatever, they were all an enormous pain in the ass and didn't work most of the time and weren't worth the effort.

CRISPR just werks

So, people are scared, because it works?

kind of.

the other big reason people care about it - in addition to making small indels, you can also use DSBs to initiate homologous recombination events. if the cell detects that there's been a double stranded break, one way to fix it is to just shove the two loose DNA ends together and fix it (which is sloppy), or you can use the other chromosome (only works in diploids or higher) as a template to fix the damage.

if you, say, cause a double strand break and also provide a template, say by throwing a pretty big chunk of DNA with some custom alterations into the cell, sometimes the cell will use that novel DNA as the template instead of the other chromosome. in lots of organisms, systems are worked out that make that procedure easy, even if it's not quite routine.

so what that means, is let's say you have a case where a human has some fucked up gene that's actually pretty important for them. you take some stem cells out of them, use CRISPR to generate a double strand break, provide a novel template that is the wild-type version of that gene, the functional one that is identical to what you'd find in any other human, and then propagate the fixed cell. once the cell has lost the plasmid containing Cas9, you have potentially a seamless edit that is literally indistinguishable from a hypothetical case where that mutation never existed in the first place. propagate that cell, transplant it back into the host, and you're done.

there's been cases where someone fixed some mutations causing beta thalassemias in patient stem cells and demonstrated that they had fixed most of the causative mutations in an undetectable way. theoretically, they could have put those stem cells back into the patient's marrow and largely alleviated their condition.

the hoopla is, yes, that now we have the means to make any edits to the genome we want in a highly reproducible and easy manner. we've had similar tech before but nobody cared because it didn't work anyway

Maybe you can confirm or correct this, because I don't know what the fuck I'm talking about and I'm quoting people who probably didn't know what the fuck they were talking about either. But I heard that some geneticists were talking seriously about somehow using a virus carrying CAS9 to eradicate mosquitos (because omg Zika), and that was one of the biggest causes of alarm; people were eagerly talking about doing dramatic world-altering shit that we can't possibly predict the consequences of.

Is that fantasy?

silly mosquito poster you can't fool us

yeah that could probably work. i'm a plant biologist, not an immunologist or whatever you call people who study mosquitos, but i do know there's viruses used in gene therapy that can fit fairly big genomes inside them.

it would necessitate a couple steps:
A) finding a virus that infects mosquitos (preferably preferentially)
A.1) that you can modify reliably
B) finding an absolutely critical gene in mosquitos that has a compatible target site for the gRNA
B.1) again, it should probably be one that is specific to mosquitos and not any other species your carrier virus might infect that you don't want to harm
C) generating your plasmid with Cas9 and your gRNA, packaging them into a virus
D) releasing virus into the wild

probably... a couple years work, assuming the stars align? that's absolute minimum. couple hundred thousand dollars of money for the researchers' salary and materials. if there's lots of trouble, think 5-10 years and a couple million bucks. but it's a question of how much money would you want to throw at it, not whether it's possible in the first place.

i mean, i'm not an ecologist so i dont know if releasing a killer virus into mosquito populations would actually work. it's also possible that it's TOO deadly and your virus never propagates in wild populations. who knows, ~it's a mystery to me~. but the molecular biology is sound.

i mean, if you want some scary fucking spooky shit, consider this: what if someone threw Cas9 and a gRNA targeting Xist or SRY into a virus that infects the testes?

all those same caveats i was mentioning before about time and cost would still apply, except it would be even easier, because all the research legwork is already done in terms of genomic information and molecular biology of the virus and the host. it's just cloning and materials, and someone doing that wouldn't care about side effects. they could probably have their product in a year or two, if not less.

What he is referring to is a method of eradicating or limiting infectious mosquito populations (INDIVIDUAL species, not the whole genus, no ecological catastrophe to be afraid of). Scientists were trying to use targeted cutting enzymes (CRISPR is much more accurate and successful so it is what will be used now) in MALE mosquitoes to destroy the X chromosome from their SPERM (males can either give X or Y gamete to the female, X gives female and Y gives male). Since X is eradicated from the male sperm, only Y is given thus any modified male mosquito that mates with a female mosquito will give MALE offspring. The key point here is that only FEMALE mosquitoes need to feed on our blood and thus only females are potential vectors for diseases like Malaria and Zika to humans. If only males are produced in a mosquito population, the modified male offspring will mate with more females, producing more males, slowly reducing female population or even crashing the whole mosquito population with no survivors since no females, if enough modified males are released.

>SRY
lel, didn't even think about that, that's pretty sick man

...

Wrong, viruses effectively achieve the same effect for gene therapy.

True, you're right. I was thinking more from a "how is it done in a research setting right now" perspective, but yes the final therapy would probably be done organism-wide with a virus.

Bump

Time to put a dormant std in one of these lol

It's basically nothing new, we've been able to alter genetics for a long time. It's just a fancy and efficient new way of doing it that greatly reduces the margin for error and makes it much easier. It's like when cell phones came out, it's not like anyone invented the phone again, just a better way to use it.

>Elucidate CRISPR to a physics-fag

Learn chemistry, macro-faggot.

Valance shell charge FTW.

Chemists don't know about CRISPR faggot, CRISPR is all about genetics (molecular genetics and bioinformatics) and microbiology (it was first discovered in bacterium as an immunity system) , because the sequences (3'-ATGTGCTATC-5') are important, not only the molecular properties.

>Valence bond shit
DIE

Or you will surely die.

Will these new delicious CRISPs make it easier to map genome to function? How much easier will it be to make glow in the dark tomatoes?