 Hello, my name is Anthony Brottodo, I'm from Rennes in France, and today I'm going to show you how you can use Apollo directly within Galaxy to help you perform manual curation of genome annotation. So why do you want to do this? In fact, you can perform using Galaxy the assembly of a new genome and then annotate it using tools like Procaro Maker. You can do all these following Galaxy training network tutorials. That works, but the automatic annotation step is never perfect. In fact, the results you get often contain some errors in the gene structures, for example, you can get some incomplete genes, or you can get some genes that are fused instead of being two separate genes, they are detected as a single one by the software. Or you can get splity gene, for example. You might also want to add some functional annotation to the genes to give them a proper name or to assign them a specific function. So Apollo is a software web portal that will allow you to do this. In fact, it's called a bit Google Docs of annotation, which means it can be used by many people all over the world to edit annotations and every people, every single user will see the changes made in real time by other people. So it's a very collaborative way to improve genome annotation. And so this works directly from within Galaxy. Now we will see today with a specific example, we will improve the annotation of bacteria. So it will be E. coli. And you can do it, for example, after running the PROCA tutorial on this genome. Just one note on Apollo and Galaxy, the two work together only on a selected list of server. So we will use useGalaxy.eu for this tutorial. And it won't work on useGalaxy.org or the Australian server. And that's it. Before starting the tutorial, I just wanted to let you know that there is a group named Galaxy Genome Annotation at this website, which is working on all the annotation tools and the link between Galaxy and Apollo on the technical side, how the two applications can interact. So if you are interested, you can go to this website and contact us there. We also have a GitHub channel here that you can open to discuss with us. So now let's go to useGalaxy.eu. I am connected with my email address here. And we are going to find the tutorial just by clicking here. And now we are going to the genome annotation section. We have the maker and PROCA tutorials for automatic annotation. And what we're doing now is the refining genome annotation with Apollo. Let's go to the tutorial. And this is all we're going to do today. Okay, so I'll let you read the introduction. And now we are going to perform all the steps and starting with the data upload. Let's go with it. So the first thing to do is to click on the upload data. After creating a new story, we can rename the history to be more clear and to name it Apollo. And then we have a collection of data available to run the tutorial. And as is written in the tutorial, you can use the rule-based uploader here to upload all this data. So copy all the list of files from the tutorial and paste it here. That's it. Now we click on build. And we are going to add a few rules. Why are we doing this instead of using the regular upload box? That's because we want to give some pretty names to each data file that will be downloaded and to give it the proper data type. So it's much faster to just copy the list with everything filled as it should be and use this rule-based uploader to do the magic for us. So we will add the three rules to define what each column A, B, and C will do. We click here, add modify column definition, and we add three definitions. The first one is URL in column A. Then we have the name in column B. And we have the data type in column C. We apply this and we upload it. And as you can see, the data sets will appear in your history. You can close this and just wait a little for the data set to become green. Okay, now the upload is finished. Now what can we do with all this data and what is all this data? There are a few important data sets here. The first one is the seven genome. This is the sequence of a subset of the genome of E. coli of the K-12 strain. We only run our tutorial on the support of the genome to make it faster to run tools, but it works. It would work on the whole genome if you want. And we also have an annotation that was performed using Augustus of our genome sequence. That's just another tool that can predict annotation on the genome. We didn't run Procar or Maker, we used on the Augustus. And we have a few other tracks that were aligned on the genome and we will see it later in the tutorial. So what can we do with all this now? In fact, to use Apollo, the principle is always the same. To do is just an extension of another software, which is named Jbrows. And we will first, in this tutorial, create a Jbrows instance to visualize the tracks that are in history and then use this Jbrows data set to load it into Apollo. So if you follow the tutorial, you just need now to run the Jbrows tool, which is in the left menu here. And that's quite a bit complicated form to feel, but it's very powerful. So once you've understood the logic, you can reuse it to visualize a lot of different data using a genome browser like Jbrows. So let's choose first which genome we want to visualize. It is in history, the data set number seven, so we select it here, there's only one anyway. And now we just want to view a minimal Jbrows, that's okay. We just need to change, this is a prokaryotic genome, we need to change the genetic code which is used by Jbrows and we would select this one, the bacterial one. And now we have to fill all the different tracks that we want to be displayed in Jbrows. So first, we add the first group as we insert groups of tracks together. So we have one first group which is named GeneCos, in which we add a single and two tracks that are of this type, GFF, GFF3. So Augustus and NCBI are not writer genes. And here in the Jbrows styling options, just a little change to make it prettier in the Jbrows view, you see it a bit later. That's it. Now we need to add another track group where we will add some tracks from sequencing, genomic sequencing of the genome. So as usual, we give a name to the group and we select which files we want to display in these groups. So we will take both PWMM mapping, which are mapping of reads from genomic DNA from two strains, K12 and O104. Okay. And now we will add other tracks, but this time as big week XY. And we will select once again PWMM coverage. So these mapping files are to store the location of each read position on the genome so you can see it in Jbrows. And coverage is the same data, but instead of having the position of each read, we just have a coverage plot all along the sequence of the genome. It's a much more lightweight to display and sufficient in many cases. We select this option and this one. Okay. Now another group for RNA-seq data. So now the tracks, it's a BAM somebody files from top hat. So we downloaded some FastQ files from SRA with these identifiers and they were aligned on the genome using top hat. So we select both tracks and as we did just before for the genomic DNA sequencing, we have some coverage tracks, which are their top hat coverage. And we want to see the XY plot and some variance band. It's just settings for choosing how the data is displayed in Jbrows. Now we have another group of data, which is variation and we add the corresponding track, which are in VCF format. So here we have the K12 and O104 variance tracks. So this is a list of variants like SNPs or insertion or deletion that were found in several samples while aligning on our reference genome. So we want to display these variations. And the final group is similarity where we will select the remaining tracks. The first one is an alignment of some sequence from a specific variant, H4, which is a specific pathogenic variant. And we insert a few of the tracks in BLAST XML format like this. Now it's just a single one. BLAST B, this is with protein. Here we just perform the BLAST of the genomic sequence against a big data bank of protein sequences, which is SwissProt. And this is the result of this alignment and we want to show it into Jbrows. So this was run against these genes. So in fact, we took the sequences from these genes along the genomes to run the BLAST. And we want to change this setting. And it was protein BLAST search. As SwissProt is only a data bank of protein sequences. And that's it. So hopefully you have not forgotten any option in this very, very long form. Anyway, you can always rerun it and change the setting afterwards if you see that you've made some mistakes. So it is a bit long to run for the first time. But as you can see, it's very powerful because for any genome that you analyze with Galaxy, you can add as many tracks and as many different kinds of visualization directly from Galaxy and even include it inside a big workflow of analysis as a result, for example. So we just need now to wait for it to finish. OK, now it's green, it's finished. Let's see how it looks. Just click on the I here and Jbrows should display within Galaxy. As you can see, the interface is quite packed. So if you want to see it a bit better, you can open it in a new tab just by clicking in the middle mouse here. Here it is. It's much more easy to read. So you can see the different tracks that we have selected groups in different groups like this, different categories. So we can see the annotation that was performed with Augustus along the genome. So to to move along the genome, this represent the sequence of the genome. We are viewing this portion of the genome in red here. And this is the genes that were identified on this genome between position 40,000 to 60,000. OK. You can zoom in like this to see the detail of a specific gene and zoom in again. In this case, you will see colors that correspond to to the sequence of the genomes. So the center is a sequence of the genome and the forward strain. And then in the reverse strain and here you have the three translation in the three open reading frame of this sequence and the gene is here. OK, this is good. We can see a bit of RNA seek data, for example, this one. So here you can see that there are a lot of reads that were aligned. You can zoom into this region. That's it. And you can see some variants, too. So you can see here that in this position, there were some substitutions of single poly single nucleotide or multiple nucleotides. And you can see the result of last piece. So for example, here you have a gene that was predicted by Augustus in this position. And if you look closely, you have some results of BlastP versus SwissProt that were aligned in the same position. And you can see the details of which protein were aligned in this position. So this is good. This is cool. Jbras allow you to visualize this data. You can share it with anyone you want. And now we can use it for Apollo. Of course, if you want to explore more the option of Jbras, there is a specific tutorial on the DGN. Now we want to use the Apollo tools. And if you look in the list, there is an Apollo category here. Apollo. And there is a specific tool, which is named Creator of a Bed Organism. So in Apollo, you have the concept of organism. Each organism is a species, a genome that you want to annotate alone or with some colleagues. We'll see how later. So if you click on it, it should appear. It can take a few seconds to appear as it's using the API of Apollo to know which genome is already inside and sometimes take some time. OK, so now we have the list. So in this tool, you first select which Jbras data set you want to to load into Apollo. So we only have this one. We selected it. And then we have to choose which organism we want to load it into Apollo. So if you already have done some things in Apollo, you might have a list here of your your own organisms, but now we are creating our first one. So we will make a direct entry here. Give it a common name. So for example, we will call it E. Coli key 12. Then we need to give it the genus. Shisha and the species. Coli. OK, and then we have to specify if this organism is public public for now. We don't want to do it. You can optionally select a specific group of users to share access to, but we have another way to do it later. So we don't care about this for now. Yes. And the other options only apply when you only want to update an organism. So for example, here we load for the first time a Jbrows data set. But if we want to to change this organism with a new Jbrows data set, for example, because we have new tracks, we can specify this option to to make sure Apollo will detect the changes. So let's try execute and wait for the genome to load into Apollo. Listen, this can take a few seconds or minutes, depending on the size of the genome. OK, so the upload of the Jbrows data set to Apollo is finished. It's green here. You can have a look at the output. It's just some JSON output saying which organism was created and some internal details for Apollo. Now, how to access Apollo? The easier is to use these two, which is annotate, open the library to Apollo. So we click on it and we just select the output of create or update organism here. Should be quite quick, hopefully. OK, this is finished. So we just click on the eye here to see. Apollo displayed inside Galaxy here. So you see that Apollo has two main parts, a small one, which looks like like a lot like Jbrows, but very packed here. And the right one with a few details specific to Apollo. So we can just resize it here to be more visible. Still, we want to see it better so we can just open it in a new tab, just like we did for Jbrows, and it should have been like that. So now we just have the Apollo interface outside Galaxy. We can also access Apollo directly from anywhere using this address, use Galaxy dot you slash Apollo. It should bring you to the Apollo interface like this. Just remember to do it after going first to use Galaxy dot you and then log in on use Galaxy dot you before going to Apollo because they share the same authentication mechanism. So we stick here and we are ready for the next part of the tutorial, which is using Apollo to perform an operation. OK, so now let's see how we can use this Apollo interface to to curate genes. The first thing we need to to look at is this track in yellow, which was not in the Jbrows data set. This is the user created annotation tracks. So this is where you will be able to add genes and to make modification to these genes. For now, it's empty. As you can see, we are we have the whole genome displayed. We will start by displaying some additional tracks that are available in the tracks tab here. So as you can see, they are they have the same categories as in Jbrows and in gene under gene calls, there are the two annotation that we already have. So we can enable both of them at once by clicking here. Or just unselect a single one or select it from here. OK, and we have all the RNA say seek similarity, etc. available below. So we focus on the Augustus track because that's the annotation that we have generated so far for this genome and that we want to improve. And now we'll focus on this on the specific region of the genome. So we just enter the coordinates here. We want to look at the region from position one to 10,000. Here we go. So I just entered it and type the enter key. So if you look at the genes that were predicted by Augustus, you can see they have some unique names that are not very user friendly. They are just based on the name of the chromosome with a with a G something dot T1 at the end. The G something is the unique ID of the chain. Now we want to see if these predictions by Augustus look look correct or not. The easiest thing to do is to compare them to results from blast against Swiss product. So we have this in the list of tracks under the similarity code and just activate this track like this. And now you can see the genes predicted by Augustus and the arguments against Swiss product of the genome. And now you can see that these two tracks are very similar because they have a lot of blast eats with the same position as the predicted genes. And if you pass your mouse above them, you can see the name of the protein in Swiss product that was aligned on a genome at this position. OK, so this region looks good, but there is another region that might be interesting for you. So we will go to position 55,000 to 63,000 like this. And now you can see some genes that are very similar to blast eats. And here in this particular region, Augustus didn't see any gene, but there are two blast eats of specific proteins that were found in Swiss product. So that means Augustus didn't predict what it was supposed to predict here. And this is what we can improve using Apollo. So now we will just try to copy these genes to the user created annotation track here. So to do this, just right click on the gene on the blasted and click on create new annotation and we select gene here. We do it for the first one, it should appear. Yeah, it just appeared in the yellow track at the top. And now we do the same for the other gene here. That's it. It should appear. Yes, that's it. There's just a little delay to communicate with the server. But now that it has appeared here, it should have appeared to any other user. Apollo or any other user connected to Apollo. So this kind of error happens when the Apollo server has a problem or if you lost your connection to internet. So I don't know what's happening. You just need to click on OK and should read out correctly. It shouldn't happen to often this kind of error. OK, so now what can we do with this two gene we have added? You can just right click on it for this one, for example. And you can see there is a huge menu here that show you everything you can do with Apollo on this gene. So you can get some information on it. You can change the type of feature it is. So we have selected gene until now, but it could be something else. And you can do a very. Different things like sitting, translation starts, modifying the ORF, changing the stop codon, moving to opposite strength. This is all options that we will not see today, but that are properly documented in the Apollo documentation. What we can do today is taking an open annotation here. And this will open a dialogue in the left in the right part of the screen here. So here we have clicked on this gene. So here we have a list of gene and we see the gene here, which is named G55. It's a type gene and under it you have the mRNA. So in Apollo, you have the notion of gene and mRNA, which means for a single gene in eukaryotic, for example, you have a lot of different isoforms. And each isoform is as its own line here and can be modified independently of others. So what we want to do now is to give a proper name to this gene because B2G55OO is not very user-friendly. But let's see which name we can give first. So this gene comes from a blasted. If we click on it here, there's a details of the alignments, the blast result that was computed on this genome. And you can see that the sequence that was lying against the genome is putative and uncharacterized protein YAPP. So we will copy this, close this dialog, and change the name here. That's it. I click here and that's it. The name is changed here. We can add a symbol if you want, YAPP, OK. And now we can sync the name to the transcript level because until now we have modified only the gene. If you look at the mRNA here, you can see that the name is still the old one. So we go back to the gene here and we click on sync name with transcript. We wait while it reloads and yeah, this is it. So now the mRNA has the proper name. And you can see in the annotation track here that the name has changed from GE55 to this new name. So now we want to do the same thing for the other gene. So here is the, there's only one, the latest gene that was modified that is displayed. We can show all the genes on this genome by clicking on show over. And you should see the other gene here. And now we want to modify its name. So once again, we look at the blasted results and we see that it's uncharacterized protein, YAPQ. We copy this. We click here to see the mRNA line. We change the name here. We add the YAPQ symbol and we sync the name with transcript. There we go. So now the two genes in mRNAs have the proper names and they are displayed here. So we have changed the name of genes, but what if we want to change the structure of genes, the limits, the start and end coordinates of the gene? So we will go to this start position of the genes just by selecting this little region. And now we will zoom like this to the base level. So what we can see is that these genes starts with this ATG, start codon here. But there's another one just a bit on the right here. So we want to just for the exercise to change the start codon to use this one instead of this one. To do this, just select the gene by clicking on it and then pass your mouth at the five prime edge of the gene, click and drag to the new position here and drop when you're at it. And that's it. Now the gene has changed, it starts here. If you unzoom like this, now you see that the limit has changed from this position to this one. We have made a few changes to these two genes and as you can see Apollo keep tracks of everything you do on each gene. If you right click on the first one and click on show history here, you can see a list of everything that was done on the gene. But the interface is a bit messed up because I have zoomed to make a more readable video using zoom, but the information is here. So we see that I added a transcript at this time. Then I changed the name and then I changed the exon boundaries of the gene. So the start position and you have the timing. You know who did what and you can see you have a preview here of the look of the gene at the different time. So if you click, you see that it was just modified at this position. And you can revert each change to come back to an older version. If for example, you accidentally broke the gene structure. You can have also the undo and redo action here to go back in time if you've made mistakes. Okay, so now we would want to add some functional annotation to our genes. Let's move to another position on the genomes, on the genome, sorry, this one. Now we have a genie here that was named G3.t1 by Augustus. But as you can see, the last results suggest that this protein might be a thrown in synthase gene. You can have a look at the alignment. That's it, name is here and the symbol is TS. So what we can do now is take this gene, drag it here to make some modification to it and modify the functional annotation. So to place the gene in the user-created annotation track, you can just click it and drag it here. That's the same thing as making a right click and creating an adaptation. Sometimes for BLAST track, for example, you can't drag it. That's why we did it like this last time. So now this gene is known by Apollo. It's in the list now here. You can follow these two lines. Once again, the interface seems messed up just because I zoomed for proper display in the video. So if you look at this one, you can of course change the name, the symbol, et cetera, as we did for the other genes. We can do it now with by tapping 309, some tabs like this and a symbol TS. We sync the name with transcript and that's it. Now it's named with a beautiful name readable by humans. Now we would look at the other tabs here concerning this gene. So let's look by the end first with the attributes. So here you can add some tag value pairs of whatever you want. So sometimes you might have some canned values and tags for very common things you want to add, but you could have anything like a source last if you want and it appears here and you can delete it very easily this way. You can add some free text comments. So I like this gene very much. It will be stored in Apollo. You can add some external database references. So for example, a PubMed ID or any accession. So if you look at the BLAST result here, you can see that the BLAST was, the aligned protein was this one with a specific ID here, P16120.1 here. And this is a Uniprot ID. So if you search for it here using Google or just by going to Uniprot and typing it here, you will have the full description of this protein in this database. So we can add this reference to as a DBXref in Apollo. So we will write Uniprot and the accession and add it here. Okay, so one interesting thing is the gene ontology tab here. The gene ontology is a vocabulary very controlled to describe molecular function. Cellular components or biological process of genes. So this way we can write into Apollo database some standardized terms that will describe what this gene is doing in real life. You can see that there is a big form here to add a new term. There is guidance here if you want to understand better what we are doing. The question now is which term we can add. So if you look back at Uniprot, you see that at this position, this gene was blasted and was matched on the genome. And if you look at this few lines, you can see that this gene has three gene ontology terms associated with it. First two molecular function, pyridoxal phosphate binding, 309 synthase activity, and one biological process which is 309 biosynthetic process. So we will want to add these terms to our gene in Apollo. Okay, so to do this, we will go back to Apollo here. So to add the gene ontology term, the first thing to choose is BPMF or CC. So BP is biological process, MF is molecular function and CC is cellular compartment component, sorry. Let's try to add the first molecular function term here. And if we go back to Uniprot, we will want to add this one, pyridoxal phosphate binding. So we can just write the name here, pyridoxal phosphate binding and you see that it's proposed. And the unique idea of this term is this one. So we just select it this way. Okay, so now we need to choose what is the relation between the gene and this term. So the most under one is edibles, which means the genes that we are annotating, edibles is like this molecular function. And now we have to tell how we can, we determine as this term should be applied to this gene. So that's the evidence code. So here we have, we can write sequence similarity because that's what we did. We blasted some genes from SwissProt against our genome and it tells us that this position, there's a gene with this function. So there's this evidence code, which is 250 sequence similarity evidence used in manual assertion. That's exactly what we've done. And now there's the last part of the form to field, which is we have some similarity. So we need to enter with what there's similarity. So once again, we write the Uniprot ID, which is this one. Okay, click on add, don't forget this. There's a reference we need to write to. So we will use it once again, Uniprot ID. It could be a publication ID, for example. If you've not, if you've seen in the literature that this, that there is a high similarity between this gene and Uniprot, but here we just used blast. And you can have some free text to comment on this. So we can write strong similarity type of them. And we can save now. And as you can see now, there's a good term associated to achieve. So everything that we have entered is now stored inside the Apollo, but you can export it. So if you right click on the gene and click on get GFF free, that's it. You can see the whole structure of our gene and all the functional annotation of the gene that is exported to GFF three format. So you can see the note that we have added. Who did what? The position of the gene on which chromosome and the good terms and everything we've entered the mRNA and the position. So everything is stored. And you can see the changes in history to probably oh, it looks like it's not maybe a bug. Oh yeah, in fact, it's not a bug. It's just that the functional annotation is not in history. Okay, so for this genome, we are very lucky because unlike in real life, when you want to annotate a new genome for this one, we have a very good reference. So we'll try to display it and compare what we've done with this reference annotation. So in the gene calls track, you have the NCBI annotator genes, which is the gold standard annotation for this genome, which is a very, very common one. And now you can navigate along the genome and zoom and see all the differences between our Augustus annotation, what we've done in the user creating an annotation track and the standard annotation of this genome. So for example, if you zoom to this region where we made some changes, you can see that Augustus did not find these two genes, but the NCBI reference genome had this gene in the annotation. And as we have added it to our Apollo user annotation track, we will be able to generate an annotation like this in the future. And you can see also all the standard names that were given to each name by NCBI based probably on the last results and user manual curation. So now we can move to another cool feature of Apollo, which is sequence alteration. So for now we have considered that the sequence of our genome is perfect, which means the assembly went very well and there is no errors in the sequence of our genomes. But of course, in real life, this is not always the case. So we can go to another position. Let's move to this one and let's display some other tracks from the variation group here. We'll display the K-12 variants. So here on the specific variant of this species, there were some SNPs and deletion insertion that were detected. And we'll see how we can change the sequence of the genome to reflect these changes. In our case, this is just SNPs that were found in variants of the strain of the genome, but it could also come from RNA-seq data suggesting that a specific position, there is an error in the assembly of the genome. As you see, these variants are within the limits of a gene. So we will drag this gene to the annotation track first and we will try to add this variant, this SNP, to the genome sequence. We'll zoom to it until we saw the sequence data here. So you see at this position, there's a variant saying that there's a T instead of a C on the forward strain. So we just place our mouse here and we right-click, how? Yeah, sorry. We right-click and we select create genomic substitution and we want to write that at this position, there's a T and a A on the reverse strain. And we can say that it's based on a C in this K-12 variants and we just click add after that. So now it is displayed like this and there's a little bug. I don't see. I don't understand why. I will just reload the page. It's just a display bug. I don't know why. Probably because I have recite the window. That's it. You see that there's a C that is replaced by a T. So in our case, it doesn't change the structure of our gene. But if there was a change to, for example, a stop codon or exon in transaction, you would see the differences in the shape of the gene here. And if you export the sequence or the GFF, you would see the differences. OK, so now we have seen the basic operations you can do with Apollo. There are many, many more operations you can do. You will probably see it in a future tutorial which with more complete description. But now what we want to do is to export everything we have done in the user created annotation track here into Galaxy to continue some other analysis. So to do this, you can go into Galaxy and you have a tool that is able to get back the data from Apollo into your history. And it is this one, Retrieve Data from Apollo into Galaxy. So you click on it and wait a few seconds for the organism list to appear. Now we can select which organism we want to export. So there is only one for us, but you might have other ones. And we want to, you can export only the genes from a specific chromosome if you want. Here we want to export everything. So we leave old Refsex and we execute and wait a little. So we have the results in the history now. And the most important thing is the annotations from Apollo here. If you look at it, it should appear. That's it. You get the GFF3 export of everything you've done in Apollo with the four genes we have created at the different positions. And the changes were made, the functional annotation, and everything. We even have the substitution that we have specified at this position, which means there is a t instead of c. OK, you can also get the peptide sequence, for example, CDS sequence. That's great. If you look back at Apollo, you can, in fact, export also directly from the Apollo interface by going to the Annotation tab. No, going to the Ref Sequence tag and selecting which format you want to export and which chromosome and downloading it from here. We'll not do it to avoid wasting time. Now there are two remaining tabs in the Apollo interface, which are Charing and GGA. GGA is very simple. It's just a chat from the Galaxy Genome Annotation community, so you can go there to discuss with us developers. As you can see, the Sharing tab is currently broken, so we try to repair it quick. OK, so the Sharing tab is repaired. There was a server-side problem that was solved. So if you click on this Sharing tab, you will be able to share the access to your genome and share a shackle to any other user of the Galaxy server you are using, so use galaxy.eu. So when you open this tab for the first time, what you see is what you have access to. So you have one organism that you can share with others. And you are part of one group which is named testing. That's just an old test. You may have seen this GX10948 in the organism name. This is a unique ID corresponding to your user on this Galaxy server, which means it allows to have multiple organisms with the same name for different users. So if you created an organism named E.coli.k12, this will not be in conflict with my organism, which has the same name, because this suffixes automatically. So what we want to do now is to share access to this organism to another user of usegalaxy.eu. To do this, you click on Group Management here. So the logic is to first create a new group. So we'll name it E.coli.collaboration. Just use letters and underscore, and it should work. I create the group. It takes a few seconds to create, and that's it. So you are now in this group E.coli.collaboration. And you can see, first, the organisms that are associated with this group. So for now, there are no organisms associated. But you can add one from the list below. So this we'll do by clicking the plus button here. So we need to confirm. Yes. So any user that will be in this group now will have access to this genome, which means it will be able to modify the gene in the user created annotation track. So that's what we've done. Of course, you can remove this organism later if you want to start to share it. And now we can change the group members. For now, there is only one member, which is my account here. But we can add as many users as we want. The only thing to know is that we need to know the exact email address that was used to create the account on usegalaxy.eu. So I know that I have an alternate account like this, with a different email address. There's an I here, as you see. So I will add this user in this group. And that's it. So if you look at the group members, now there are two Antonio Botodo in the list. One, which is me, that I cannot remove. And another one from atinaria.fr, which I can remove at any time. And now this user will be able to login and change the gene structure here. OK, so that's it. I think we have done everything that is in the tutorial. Of course, there are many more things you can do with Apollo. We didn't see all the option when you right-click on a gene. We could make it into a more complete tutorial later and maybe make a tutorial for eukaryotic genomes where you can change the exon and intro junction, for example. Thanks for listening.