foreign on behalf of the biochemical society and Portland Press I'm pleased to welcome you to today's webinar which is part of our biochemistry focus focus webinar series topics in the series include different research areas in the book molecular biosciences as well as practical sessions to support career development each webinar will give you the opportunity to ask questions via text and we welcome suggestions for future topics and speakers to feature in our webinar series please see the website for details um so hi my name is Catherine Joplin I'm the chair of the biochemical Society jeans panel and so I'm here to introduce today's speaker I'm an associate professor at the University of Nottingham with um an interest in RNA biology so in this webinar we're going to hear from Dr Frank Van Eden he's from the University of Sheffield and he will discuss the theoretical and practical aspects of Christopher modification in zebrafish just to introduce his work zebrafish is a widely used model organisms funding the fields of Developmental biology epigenetics transcription and more the Advent of crispr technology for making precise modifications to the genome has surpassed s-i RNA and talent Technologies in many applications much work and Outreach has focused on genome editing in mammalian systems however vertebrate model systems such as zebrafish and xenopus are used by multiple vibrant research communities who have adopted crispr technology for their own purposes Dr Frank Van Eden is a senior lecturer in the school of biosciences and his research interests includes using this upper fish as a genetic tool to study development and disease um so just before fate gets started on his webinar um to mention this questions will be asked after the presentation so in the meantime please type your questions into the question box um which you can do at any time during the talk when it gets to the end then I will try to arrange for break to answer as many questions as the time allows okay so that's it for the introduction so I'm going to hand over to Frank now who's I'm getting his slides up on the screen and I'll see you for the Q a session at the end every day Frank okay um thank you um it's it's a pleasure to be here um I'm just seeing whether this is also infused so I'm just gonna move this out of sight a little bit um so um this is the first time I do a seminar like this although we are doing these in the university and it is because um many people are working on different systems and they're um you know kind of keen to to get into the zebrafish especially since now it's so easily possible to make Knockouts in the zebrafish using a crisprick technology and um and and actually also further genome modifications although this is a bit more um you know still still a bit more uh uh work so to say it's not as well established um but it it may be good to Simply give a kind of a practical guide of how we do this because we have made mistakes and we have learned uh things along the way and I think it's um uh good to maybe share um how how we go about and I thought I'd do this um simply by kind of using a hypothetical scenario and I I hope that is a good idea so the scenario that we that that I would do is you know you have identified a human or a mouse Gene of interest and you would like to create a zebrafish knockout so what do you do next where do you really start um well the first thing you need to do is if you work in a different organism is that you have to identify the zebrafish homologues of your Gene um and then you need to design a crispr so what we use for identifying homologues is is basically out of convenience we tend to use ensemble and Ensemble is um is a website um I hope you're aware of this where you can find your Gene of Interest relatively easily so um here we went for a human gene so I I kind of chose uh the power to have one gene this is because we've used it in several practicals and um and I found the human gene uh on Ensemble as you can see over here if we want to then go to the zebrafish homologues of these it is very advisable to look at a gene tree and I will show you that in addition we we look at another place and that is a list of orthologs and these can all be found in Ensemble if you look for the gene of Interest so if we click on the gene tree what you will get is a tree that roughly looks like this and you can navigate this by clicking on some of the buttons over here and that will allow you to find uh you know the homologues of the of the human gene which is located over here and you can see it nicely in a phylogenetic tree with the zebrafish Gene and this is really important because the zebrafish has underwent a partial genome duplication and often where you think there is just one gene it turns out that if you look in detail in the sequences you can see another Gene over here which doesn't have a very um you know intelligible name but nevertheless it is an ortholog of the pow 2-1 power to F1 gene in humans so if you want to create a knockout you better prepare to make a knockout for both this Gene and this Gene because um they're both part one uh they're both homologues basically if you look towards the orthologues you can also see that there is not a clear relationship between the zebrafish and the human because it says over here if you look at this list which you get if you click on there if you look at human you get a one to one or a one-to-many relationship with zebrafish so this is an important piece of advice when you start such a knockout project be very clear that you have you know that you know of all the zebrafish orthologs that might be present you might you can easily miss one okay once you have identified one of the genes that you will knock out you can simply click on this Gene and then you will get to the gene landing page for the zebrafish so it's very convenient clicking through the organisms and then if you go to the transcripts that you can see you need to select the transcript that you want to use for creating a knockout and what we usually choose is the proteins that are the longest and there's a tag over here which is very useful that says Ensemble canonical that means that this is the the most common transcript variant that is present in the zebrafish and that's usually what we use as a start and then when you want to design your Crispers you need to um you know choose your target area very wisely and you might say and and we have made some very painful uh um experiences with this you might say okay I'm really gonna make my crisper mutant really early in the gene because then we are sure that everything else behind it is going to be destroyed and knocked out um however this is often not the case if you make a a frame shift mutation really early in a gene often the ribosome simply finds the next atg molecule or the next atg codon and we'll start forming the protein again so uh so so so you have to be careful with this of course if you find if you make a crisper that is you know very very far in the protein then you you will have some Active Components being uh being being still being present and in order to look at where the important parts of the protein are we simply in Ensemble can click on the protein and this will bring up this kind of protein analysis where several conservation detection programs are used to identify the important parts of the protein and for power 2 F1 which we're using as an example over here you can clearly see there's something very conserved over here a power specific domain and a homeobox domain so it's probably going to be a transcription factor and that means that if you make your mutation your frameshift mutations before the actual active domain then you will be in business another important thing is to look at splice variants and this can also be done simply on the landing page for the gene in in in zebrafish and there you will get um you know these different transcripts which you see over here you can recognize them over here and you can see that there are certain code certain exons as you can see these ones and these ones that are specific for certain isoforms or or splice variants of the Gene and of course if you want to create a knockout you don't want to have you know this isoform knocked out but not that isoform knocked out so when you decide where you want to place your your crispr mutation try to select axons that are common to all of the splice variants that are shown so I would normally go for you know X on three or four in this in this case right so where do you go next Once you have identified the transcript that you want to use to identify your crisprs and roughly the axons where you want to do this then you can take this number over here you have to remove the Dot 3 I think but you can take this number over here and then copy it into uh The Chop Shop website so this is the website that we tend to use to design our crispr guides um it is basically out of convenience we find this one the most convenient um there are a few other ones um I I you know there are other sources where you can see guides so there's a uh if if you're if you're familiar with the UCSC genome browser there's a zebrafish genomic Hub that you can find under the track hubs it's called zebrafish genomics made by the Burgess lab and if you activate this track you can see a kind of pre-designed guides that are you know according to their algorithms are are likely to be active guides however as I said we tend to use the chop chop website if you use this there are a few things that we need to pay attention to first of all you need to use the right version of The genome so there's another one which is called 10 we use 11 that's the newest version and in the basic settings we will set the product size and then we'll come back to this later um to about 90 to 175 before I actually send off this sequence right okay when you send off this sequence after a while um The Chop Chop website will uh uh will produce a view like this one over here and you can see different axons of the Gene and above this are all these little uh little arrows which are are basically designed guides and you can then scroll and zoom in on the on the axons that you were interested in so you need to make a note of that which axons you actually want to Target um and then you get a view like this and here you can see different guides that have been designed by chop chop and they have a kind of a scoring system where the green ones are the good ones the red ones are the bad ones and the orange ones are the ones in between however we have found that these scores from Chop Chop are are relatively poor predictors of in Vivo efficiency and actually we find that many um many guides are are perfectly active that have low scores so when I look at these guides I actually look at a number of things simply by eyeballing them so I I generally worry if there are long stretches of single nucleotides or dinucleotides so something like this I worry a little bit about it but you know if I have to use this guide I wouldn't worry of ordering it they're not that expensive anymore I am a bit worried about self-complementarity so for instance here's a guy that has quite a bit of self-complementarity um I I I I I I would avoid using that guide if I could um sorry then there are these tables over here this shows the number of mismatches so this is a perfect ectopic site for the guide where it would cut in a genome there's none over here but there are some guides binding sites in the genome that have more than one mismatch and you can see those over here so this is guides with three or guide binding sites with three mismatches um I generally do not worry too much about these mismatches unless they go into the hundreds and that is because um you know we are working in zebrafish zebrafish can be crossed after you've injected guides and that means that if there are of Target mutations you can clean them away from the mutation of Interest by Crossing them out to wild-type lines and zebrafish have 25 chromosomes so that means there's only a four percent chance that any of these off-target sites you see over here will actually be on the same chromosome as where you want to place your mutation and that would mean that it would be more difficult to separate so in in most cases it's very easy to get rid of any off-target mutations that you might have induced in your fish after applying the crispr cas9 okay so um this is how how a crispr looks on the genome how it's actually binding so you can see here the um the crisper um the the CR RNA here you can see the Tracer part of the of the guide RNA and you can see that the CR RNA binds to um you know here the Strand that is opposite if you read um the the gene sequence over here and here you can see the Pam site which is which is an essential sequence for the guy to work and it's basically it consists of two G's if you read in this direction now what we do if we are on the chop chop website is we we we we we just look through the guides and we find guides that disrupt the strict restriction sites and ideally we select enzymes restriction enzymes that cut directly into PCR buffer and if you want to know which ones that are you can look on this website over here if you need to use um a guide that has no restriction enzymes that is also possible I will come back to this later anyway so you click on one of these guides over here and then a chop chop will bring up the next window which is this one over here it shows the guide on the sequence and what you can see are over here are a number of primers that have been designed to allow amplification of the region where the guide is actually going to work and these are going to be handy to determine whether the guide is actually doing what it's supposed to do when you inject them into zebrafish embryos and when we evaluate a guide what we do is we look at the Restriction sites that are also annotated in this particular view you can hover over these blocks over here and see which restriction is or which restriction enzyme is actually um you know is is detected over there and that will mean um so so if you find for instance a BSL one side is a fantastic enzyme because it it has two C's over here then it has a number of undetermined nucleotides and then two G's that's the the the recognition side of BSL so it means that if you have a guide that will cut over here and will create insertions and deletions at this particular site um that um you know this will destroy the restriction site and you will be able to see this and evaluate this on a gel um you know very soon after you've done your injections foreign you can also as I said you get this list of primers that you can use to you can order these primers along and and you can use those then to evaluate whether your guide is working okay so when you've chosen a certain guide you can order them and then the next thing is basically you inject this guide into zebrafish embryos and I've put the um you know the pipetting scheme over here because I I always like it when people publish this that's very easy for me for for you to to copy so this is what we pipette together um and I will come back to some details of this later um we inject the zebrafish when the eggs are freshly laid then we wait for 24 hours we do a hot shot DNA isolation it's a very quick and dirty isolation and it only works on relatively short PCR products so that is why I said that we need to set our PCR products to a relatively short size when we go into the chop chop website what then when we have isolated this DNA we do a PCR with the primers that we have that we have ordered over here and we do a restriction digest and we can do this restriction digest and directly in the PCR buffer so everything is very convenient in this respect and then you get your gels you run your gels and you can see whether your guide is working so here you can see some uninjected embryos we do the PCR when it's not digested you get a straight band however if you're if you digest your PCR product you can see that you get a nice digestion product and you can see some different examples over here here we have embryos that have been injected with the guide now the guide cuts the DNA and as a result of this cutting the DNA repair mechanism will lead to insertions and deletions and these insertions and deletions will destroy the restriction site that we use to digest the DNA and as you can see over here um you know we have digested DNA and undigested DNA and we can see that all the PCR product there's almost no digestion taking place anymore in comparison to the undigested ones so that means that the crispr that we have selected is actually working and it's creating mutations at the site of Interest and if this looks okay then we'll take a parallel batch of fish that we haven't isolated DNA from and we raise the fish that are um that that have been injected so we raise the injected embryos these Mosaic these embryos are Mosaic so they have all kind of different mutations in that germline and what we do next is we cross these fish to Wild type fish and then we look at the embryos again so again after 24 hours we do a hot shot DNA isolation we do PCR we check whether whether the mutation is there and then we actually start sequencing the fish um the the the the the embryos that seem to have a mutation and this means that these embryos are going to be heterozygous for an induced mutation and um and they also have still a wild type allele because they have been crossed to Wild at fish and we basically sequence the heterozygous fish straight we don't do anything special there that means that if you get traces of sequencing that you get complicated traces and I've shown an example over here so this is a sequencing trace of a wild type fish and you can nicely see the sequencing Trace you can see the Peaks occurring over here however if we sequence a fish that had a mutation you get a complicated pattern and there is actually software that promises to deconvolute these patterns and tell you what the two sequences were I generally do not do this I simply do this by hand because I find that it's just as quick so what you do is you look at the sequences over here and you compare you line them up and you compare them directly and see what extra see what extra Peaks have come up and if you do this you can see that over here there's an extra T here there is an extra t here there has come an extra C here there has come an extra G sequence because um there's still there's still a single G Peak over here so that means there has been um you know the the shifted base is a g here you can see instead of a t you can now see a T and A C so there's an extra C has come up if you know have a few of these sequences read you can look at this sequence and try to find it back in the sequence over here and as you can see the TT cgc can be found back over here in this region over here so so it means that this particular sequence has shifted now to over here and you can then count back you need to six shift the sequence one two three four five bases so that would mean that this is a five base pair deletion um uh in in in the sequence so we know directly what is going on here if this is too complicated you can also go to the end of your sequence Trace since we have very short PCR products we actually read until the end of the PCR product and you can get this you know very particular um you know sequence endings in your in your sequencing traces and um since um one allele will always be longer than the other when you sequence two alleles which there are which have a frame shift it means that the messy sequence that you have over here at the very end of your PCR product the sequence will clear up and will become clean again because there's only one PCR product left at your sequencing and that means that you can actually find back patterns over here so I see a green Peak and a few red ones and if you look at the earlier sequence over here I can see the same pattern back and you can simply move um the sequence back and count back how many extra um or how many how many bases have been inserted or deleted so here you can see I'm just going to change this so here you can see in a peak the last a peak and then if you go back one two three four five you can see an a peak again that looks very similar so you see the repeat of the pattern over here so that means that we have a five base pair insertion or deletion that's now not completely sure but it's five base pairs and that's really important because five base pairs means that there is a frame shift so it's a frame shift of five it's not a multiple of three and that makes us happy because that means that you know this mutation will send the coding sequence out of frame it's a frame shift mutation and it will usually cause a null so that's what we do we sequence the the the progeny of such a cross and then we uh we we raise that fish we raise a parallel batch or you can actually do tail clips and then identify the right fish and then we raise these fish we re-identify them you know we make sure that they have indeed the right mutation and then we cross two of these different uh fish together um to actually get a no phenotype so that's how we how we make uh no mutations uh in the zebrafish now there are a number of other techniques and I'm not sure how far I am I still have some time so I um there's there's a a technique called crispr eye where you rather than um you know making a mutation or making a null mutation you try to block transcription I must say that we have spent quite a bit of time with this and we find that this is not very efficient we have not had had very much success with this and so I I would at the moment I would not recommend using it unless there has been a new paper where they showed that this technology is you know they have uh strongly improved this technology another technology is using crispants this works quite well so what you can do is you can if you have a very efficient guide is you can um you know identify mutants by directly injecting guides into um the zebrafish embryo um usually if you inject a single guide it is unlikely to suffice and it's a there's a statistical reason for this and I don't have time to go into detail for this but it means it's best if you want to use this technology is to select two to three different axons and design a crispr against every different axon and then inject them together to create mutants to create the crisprints so where you knock out a gene directly in the in the embryos that you have that you are injecting you have to be aware that if you want to publish such a method that the reviewers might start asking you know it's very nice that you have crispants but show us that you also can make a mutant that shows the same effect so you have to be a bit careful and I would always use two these two approaches in parallel so here you can see an example where I have created a double crisp and so we made a crisp and for two different genes this is an uninjected set of fish here we injected two patched one guides and they have a very slight phenotype it's hard to see but I I can kind of pick it out these fish were injected with two patch two guides they're a bit more affected still they look quite normal but then if you inject a combination of two patched one and Patch two guys we create you know a double mutant or here actually a double crisp and for both patch one and Patch two and you can see that these embryos are quite severely affected and here you can see for comparison a real patched one two Mutant and you can see that these embryos look pretty similar so we we did a pretty good job in in knocking out both jeans right now another technique is Gene editing and this is um this is actually something that is that many people are very keen on so this is rather than making a simple frame shift mutation to create a non-mutant um rather you want to make precise point mutations and um we we have been using this and we have had some success with this it's still uh you know it's still not super efficient but it is something where you can really say okay this could be a good PhD project to start trying this um what you need to realize then is you know how to do this the first thing is if you want to make a single point mutation um the guide that makes a cut in order to do this needs to be very close to where you want to make that particular mutation it should not be further away than 10 bases of your target area and I found this um this efficiency uh graph from the IDT website I think where they show this so here they try to make mutations um either on the cut side or further and further away from the guide cut side and um you can see that if you if you're very close it's efficient but if you go more than 10 base pairs away the efficiency drops off very quickly okay so if you can only design or if you can only make a crisper about 10 base pairs away from your target nucleotide that you want to change um then um it's it's you know you're very restrained in your crispr choice you can only take you know a few crisprs will will actually be near enough to the to the mutation that you want to make now if this is the case as I said if you get uh if you find a crisper nearby and it scores quite low don't let these scores demoralize you it might still work um if you really cannot find a crispr then there are other systems there for instance cpf1 where the pen has a different uh recognition site maybe you can find a cpf cpf Target site near the nucleotide that you want to change um of course if you have very few choices for your Crispers it might not be possible to find the restriction site that gets you know destroyed or damaged by an active uh by an active Christopher cast 9 complex so that means that you need to have a different test to see whether your crispr is actually active and what we use there is a very simple test it's we call it the smear test um it's very low Tech the the only thing that we do is we inject our fish with the crisper and then we do we make a very short PCR product so again we try to get it about 120 base pairs that is really really short and then we compare injected embryos with uninjected embryos and you can see an uninjected embryo over here and your PCR product is really nice and sharp you get a crisp band however if you look at the injected embryos you can see that the bands are all very smeary and hazy and that is because the crisper is working it's creating all kind of um you know uh base insertions and deletions and that will cause um you know PCR sizes to be different so you get a very you know imprecise PCR product that is you know it's it's covering a range of sizes so that means that we know yes the crisper the guide that we're using is efficient and is working if you don't want to use this technology there's another technology called crispr stat and I've put the um the the website uh for that over here okay um so if you want to make such point mutations you need to make um uh I didn't say this you need to create um uh basically a primer or a piece of DNA that will guide the mutation that you have to make so what we use to actually you know uh change the the sequences we use of 127 base pairs roughly editing oligos these editing oligos have the edits that we want to have in the DNA in addition these oligos are phospho phosphorothioated because they are more stable and actually IDT appears to have a new modification and you may want to try that maybe they're even better they're called alt R modifications so we make these oligos in a particular way and I've I've tried to show this in this figure over here so these oligos are asymmetric and they're complementary to the non-target Strand okay so I showed this over here so the oligo goes in this direction if we say that we want to make the edit over here then there should be 42 base pairs on this side and there should be 85 base pairs on that side and and and this is the paper that that that basically says these oligos are the most efficient and this is only forecast nine if you use another enzyme you need to design your oligos in a different way okay so once we have identified where we want to make the oligo we need to do additional things and this is really what I call the puzzle because it's really complicated but we usually get it worked out so first of all we want to identify we want to add additional verbal bases we want to add it further wobble bases in the coding sequence to ensure that we create a primer binding site and and we're going to order a corresponding primer then that will only um you know bind the the target area if the edit has been made so this is a kind of our positive control that we're going to use to see whether um you know the edit that we're trying to make is actually going to be made in the genome and for that we need another primer and this primer needs to be outside of this targeting oligo otherwise we simply start all it start amplifying the oligo that we co-inject when we design these when we change these wobble bases um we keep an eye on coding frequency I've put a link over here um and um what we also do is we add it in such a way that we try to create a new restriction site and again if we try to create a new restriction site we use enzymes that cut in PCR mix um to make our life easy finally we need to do one more thing we need to ensure that when the edit has been you know entered in the genome that um that the guide doesn't keep on cutting the edited genome because otherwise you will create additional mutations that you do not want so we add it in such a way that we for instance destroy the Pam side that there's no cutting after the cut the first cut has been made and the edit has been done then the the guide should not work on the genome anymore so here's an example for a particular mutation that we're trying to make I actually do this old-fashioned and do this in word um I use um I I have a whole load of um a different um colors over here so if we want to make a d2998a mutation so we change the D to an alanine um then this is the codon that we want to change if we want to change to an Ln in we have a number of options we have a GCT GCC GCA or gcg and this is the codon frequency this one is very rare so I'd rather not make that particular change in in the genome what you can see in yellow is the guide sequence and in Gray you can see the Pam side but it's reverse complement so it's not a ngg but it's CCN then in bold you can see the target codon as I said and we have some alternative bases over here for wobble sides and you can see that you know the ones in red I'd rather not use but I will use them as I have to and then this uh particular nucleotide is very close to a splice acceptor site I'd rather not change any nucleotides that are close to this you know that are part of this splice um sequence but outside the splice sequence I can make a few changes and and and here I I I I edited these two uh these three bases over here because this is not coding sequence I can basically change it to whatever I want right okay so then I simply I try different versions of changing these nucleotides and then I see if primer three um you know is capable of identifying a primer that that will bind to mutated sequence so here we have the modified sequence when I made all the changes and if you look closely over here you can see a t c t a g a so this has introduced a new um uh Xbox One restriction site in just you know outside the coding sequence in the intro now we can use this to identify um you know the edited allele I also have identified a mute um a primer binding site that will bind over here it's underlined over here and you can see that this primer will have a lot of mutations compared to the original sequence so all the mutated bases are shown in green over here and this will then allow us to design two primers that can amplify only if the edit has happened in the genome so what are we injecting so we inject the Tracer the guide IDT says it's good to pre-form these molecules um so um so so so so you keep them for uh for a few minutes at 95 degrees and let them cool down then you add the the the the the cas9 protein you add some phenol red that's simply a tracer to show where if you inject it and then you keep them at 37 for a few minutes then when the active press broadcast 9 complex has formed you add the editing primer and this is then what we inject into the zebrafish again and then we do a PCR test for the editing after 24 hours and this is what you hopefully should get so here are embryos that have been injected with the editing mix this is an embryo that has not been injected with the editing Mist mix and this is just a water control and you can see that only the embryos that have been injected with the editing mix um you know will allow amplification uh with the added specific primer we can then also do um you know we can use primers that flank the editing site and then use the new restriction enzyme to get an idea are we inducing any new restriction sites and here we have just a whole load of embryos that have been injected and yes we see a few embryos over here where suddenly we get a smaller product after we um after we cut with this enzyme so there seems to be some editing going on here and that's that's very encouraging what we do then is we raise injected embryos these embryos are Mosaic so then we will have to identify we have to cross the G zeros and then you know look at individual embryos that are progeny from such a fish to see where we find any fish that have the edit and this is it can be seen over here so here we have a panel of F1 embryos from a particular founder fish and you can see that three of these embryos when we use the um you know the edit specific primer we see um you know three embryos actually yes they get an amplification and if we see that we can then take those particular embryos we then amplify with the flanking primers again and then we do sequencing of um of of of of the DNA this DNA is then again going to be heterozygous because we crossed our G zero fish to a wild type fish so we have a wild type chromosome and we have a mutant chromosome and you can see a sequence Trace over here or you can very nicely see that you know um in in in in in in two bases where we wanted to have the edit we can certainly see we have double Peaks so that means that the fish are heterozygous for the mutation that we want and this is how we you know how we are currently now you know creating very specific disease models where we uh you know choose exactly one amino acid that we want to change into another amino acid and then we make that mutation in the zebrafish okay I think I'm getting close to the end of the seminar I'm gonna skip this um if you want to make more um you know larger insertions larger changes of of the genome we haven't used this um at the moment where we're gearing up to this if you want to do this there's a a very good online protocol um it's shown over here I'm just going to get the thing out of the way so this web this publication over here gives you a very detailed protocol on how you might do this um and I I can truly recommend this we're actually going gearing up for this now I'm gonna stop here I just want to acknowledge the people who have been working on this and I'm I'm ready to take questions now so I'm not sure what I need to do now yes um you just sit there and I will read out the questions so thank you very much fake for a really nice it's very nice to see such a kind of detailed idea about actually how to do it and I hope it's been interesting to the audience um so as a reminder please just go ahead and type questions into the box that you should be able to see on the go to webinar control panel and um I will read them out for Frank to answer um so as a starting point um I guess I was interested in what sort of proportion of HDR events versus sort of background nhcj deletion type things do you get when you try to make those point mutations okay so that's very very um oligo and and crispr dependent so [Music] um I think we we need to raise about I I we once had some statistic we need to raise about 36 G zeros um to get a transmitter that sounds like a lot but it's for fish you know that's one tank yeah so in that sense it's quite doable um but it's so and this is an average so some of them you know within 10 fish you will have one that transmits the allele that you want but in other ones it was you know 60 fish before before people found the fish that they wanted so um it is and we think it is dependent on you know how efficient your guide is so it's really important to test um you know how how well is my guide actually making mutations if it's very poor at that then I wouldn't I would probably not advise using it but there's probably also factor x which is you know the the the oligo that you're using some oligos we find are very efficient other oligos we don't we don't you know they're not very efficient and we don't you know we don't get the statistics there we don't have the money to try a hundred oligos and see you know what the sequence record requirements for this are so we we generally um you know just try injecting and as I said if you um of course you can do these very quick tests after 24 hours after you've injected your oligo and crispr guide combination you can do your pcrs if you see a lot of bands coming up that were not present in your uninjected embryos then you know yes something is happening and it's probably okay yeah so yeah that you see a lot of negatives in your in your injected embryos where you don't see any amplification then I would say yeah probably the you know try something else try another guide or try another oligo to uh to to make your mutations we have had at least five successes now in the department where people did that and it's maybe out of trying uh 11 or so so it's yeah it's about 50 success rate and usually okay you can you can discard the unsuccessful ones quite quickly because they don't give these pcrs so there's not much time lost it's not that you have to raise you know the unsuccessful ones where usually weeded out in in the very early stages and not you know after after three months raising adults and trying to find a mutant and then finding that it didn't work so the cost is quite Limited in that respect sure yeah so it's not like you have a particularly strong feeling of whether HDR is more or less relatively efficient than it might be in a different organism it's the same as the principle of you just need to do it on a fairly large scale and it's it's I think uh you know uh any any and what's called a non-homologous and joining is probably more efficient I'm quite sure of that yeah um so it's it's it's probably the you know it's 20 I would say uh you know of the alleles will be will be uh you know driven by the by the oligo and and 80 will be simply insertion deletions yeah yeah so so I think that would be realistic unfortunately there are also some hybrid things so sometimes you get your mutation but then you also get an insertion or a deletion and that's of course not what you want so even if if you edit your allele such that the guide stops working you will get a few dirty alleles so to say sure yeah so we've got three questions in the chat at the moment um the first one is from rather Kulkarni and um she says thank you for the talk you said that the chop chop scores did not reflect the guide efficiency in this case how do we select guides for Chris Vance where we that need very efficient guides okay so the good news is many guides work very well if you simply uh you know use them by ordering the the RNA and the protein 75 of the guides work you know if you randomly pick a guide even if you don't even look at the scores at all I would say 75 work to a very decent degree so you could even simply go and say I'm just gonna you know look for a GG somewhere and then I'm I'll I'll I'll I'll I'll take it from there so the only thing that a guide needs is a Pam site right so so I I if you if you if you order in protein if you order in the RNA and not make it yourself it is very easy to get an efficient guide it's it's really not hard so um so I would I would I would still have enough I would have an I would look at the scores but I would also say okay I'd rather have you know one guide in Exxon one one guide in Exon two and one guide in Exon three and I would simply take you know the best scoring or if you if you if you say you know I really want to see whether they're working then pick one that has a restriction site close to where the guide actually cuts and that it will you know lead to destruction of the Restriction side yeah second question from male Bellic um it says have you tried the new cast variants SPG and spry with different Pam by any chance no no we haven't died we haven't really had the need so far you know if if you really want to make one particular mutation and there's no no guide near then you know you you kind of think okay I should take another one um but we never had that issue we always found a guide that we needed you know pretty close to the mutation that we wanted and and as I said there are cpf1 which has another specificity which is very T Rich so you know GG and CC uh are are are by default if you have an at Rich region then that's not gonna you don't you don't find very many but then you have cpf1 which has the the TT TV uh Pam site which will work quite easily so and and we've tried that one and that works pretty well as well so I'm I'm quite happy with that the new variants I'm not sure how easy it is to get them because you probably would have to um they're probably I'm not sure whether they're commercially available as proteins yeah so yeah I guess a big advantages you can just have you tried different suppliers for cars 9 actually and found any variability in that so we have tried IDT and neb and both of the work work for fine so we go by Price basically and and you know laziness because we know that neb works so we and they have it in micro molar and it's easy for our calculations so we tended to have any be uh uh for for our for our and we have very good experience with that NAB is a good company so um you know that's we've got a question in two parts from Antonio Martinez cielver um first part says thank you so much for the tips I'm actually genotyping my F1 and could be difficult to find the changes in the sequence respect to Wild type and the second part says could you talk about software in development to analyze the sequences okay so so software that deconvolutes um yeah as I said um My Neighbor Next Door uses it and and I I just looked at some sequences this morning and I I felt like I can I can I can just as easy and look at it so I I haven't really um but um if if he can supply or he or she can supply an email address to me um I'm easy to find online just type in my name and you will find me send me an email and I can I can uh inquire next door to see what kind of software they're using because it looked pretty nice thank you that's really happy I Anna Lopez Ramirez how do you see QR with primers targeting the last sex signs in the gene to evaluate RNA nonsense mediated decay of the specific Gene after crispr injection uh we haven't really we haven't really um uh done much qpcr on a crispr injected ones we usually then go through the genome and or sorry go through the generations and generate the mutants before we do any qpcr and then the the options are you know you can use basically any qpcr primer set to do that and we usually use gen script to design they have a website where where you can kind of design your own uh um uh PCR primers yeah and just had a comment here I guess people online can't see the questions so I just read this out um Jordan Walker says that Chris ID is a good de convoluter so there's a okay um two more quick questions then we better wrap it up um so one of them some another Kushner tantry that says have you tried Chris Vance if yes what was the efficiency you found so so making Christmas works pretty well as but you it's best to pre-select your guide so when we do this we order four so we we pick four using chop chop and we spread them out over different uh axons ideally and and and then we inject all four of them and we do this quick smear tests and then we look at the guides that give us good smearing um and then we pull usually two or three of those to do the final injection and and that works really really well surprisingly well much better than than morpholino's which are really expensive and have a lot of issues so I I if you tell me you know what do you want to use morpholino or or or Crispers I use crisp and all the time definitely and then one final very quick question um before I wrap things up um rather um thank you for the great talk have you tried making self-specific knockouts [Music] this has been on our mind for uh for a long time but um it requires you to to generate transgenics and and you know specific um but I I it's definitely not impossible you know it's it's definitely possible uh but you know it it will take a serious amount of time to get this working yeah so I better wrap things up now as we're almost at times so thanks very much indeed to everyone who's come today and for the questions in the chat and thanks very much for to break for a really really interesting talk and for all your responses to the questions yeah and please answer questions by email if anybody you know has a has didn't get it or it went too fast I often go too fast or it was too too basic or too advanced and just send me an email and I'll try to respond that's fantastic yeah thank you for that um I've also been reminded to tell you that you can continue the conversation online um following at by kensock or at PP publishing um we welcome suggestions for future topics and speakers to feature in this biochemistry Focus webinar series if you have an idea for a webinar in 2023 we invite you to submit a proposal for an upcoming webinar you can find more information about the webinar's proposal webinar and watch previous recordings on biochemist www.biochemistry.org events um hyphen and hyphen training um just go to the website and have a look website for all upcoming future webinars that if you have missed any of the 50 plus webinars that we've run as part of the series or would like to watch them again please visit our website or YouTube channel um so detox should be shown on the screen there hopefully um the recording from today's webinar will also be able to what available to watch within the next couple of weeks join us on 10th of March at 1400 GMT for our next webinar transitioning between Academia Academia and Industry and vice versa where we will hear from four speakers as they discuss their individual career paths and motivations as well as provide helpful tips and advice on the skills needed during the transitioning process finally I'd like to highlight that it's more important than ever to stay connected and engage with your fellow molecular biosciences it's an exciting time to join the biochemical societies community of researchers and Specialists to stay connected and take advantage of key benefits including discounted registration fees for society courses and meetings exclusive access to a wide range of Grants and bursaries personal online access to two of their journals and more um visit the website to find out more and finally I think that's all I have to say today so thanks very much again to everyone for coming and thanks to Frank for a really really interesting webinar and goodbye everybody bye
2023-03-23