Okay everyone, it's time to do the final part of our webinar on the codon tree pipeline in PATRIC and that's how to find out more about it, especially if you're having trouble or questions. The second is how to cite, as in how when you're writing a publication for your tree, how you can write the materials and methods. I've got it all ready for you so that you can use that. Let's go first to our tree job again, I click on jobs. It overwrites the page. Let's choose a job that I'm unhappy with. Although this isn't a tree job, this job failed. I can highlight it and if you have a tree job that fails, we want to know about it. How about this one? Maybe I'd think it's been running too long. It's been days. Generally, the trees will take days. But let's say it has taken days and you're concerned, you can click on the report issue icon, you can also click on my kill job. But just go ahead and report the issue to us. I click on that. It opens up a window that gives me some information and you can see something like either my tree job failed or this job has been queued for 48 hours and then you would submit it here and that would let us know right away. That actually comes directly to me. So I can look at it, I can figure out what's going on and ask if we need more servers online or something like that. Look, if a job fails, we don't know unless you tell us. If you're unhappy with something, we don't know unless you tell us, we want you to be happy. We want you to be happy. We want us to be happy and the way we are all happy is that if your analysis goes very smoothly and you publish in really great journals and you say you did your analysis in PATRIC, that way, you're happy. We're happy. When something goes wrong. Tell us about it and we'll try to work to fix it for you. That's how you submit a problem when you are having a problem so that we know about it. But let's go to our successful tree job. We view it. Remember we opened a wonderful report and we just loved it and we love the tree that we saw. But a question you might have as well. Yeah, it's great, but how was this tree built? Right to say I used the codon tree pipeline in PATRIC, not exactly. Close, but not quite. I had another PATRIC app open here. Let's look at the services and when I go to the phylogenetic tree, and up here, and also here are links to the tutorials. You're going to really appreciate seeing this, especially when it comes time to publishing on your tree. Let's click the tutorial here. It'll open up a new tab that is on the codon trees and this is where it shows you how to step through step-by-step on submitting the tree job and interpreting it. Going to increase it here because this is what you should use. Not this sentence because we've retired the first sentence, but this one. It uses PATRIC's global families. I would say we use the codon trees pipeline in PATRIC which uses the amino acids and nucleotide sequences from PATRIC's global families which are picked randomly to build the alignment. Then this part, both the protein amino acid and gene nucleotide sequence are used for each of the selected genes from the PG Fams. Protein sequences are assigned using muscle and at the bottom of this tutorial are the references. The nucleotide gene coding sequence are aligned using the codon aligned function of BioPython. A concatenated alignment of all proteins and nucleotides were written to a Phylip formatted file and then a partitions file for RaxML is generated describing the alignment in terms of proteins and then the first, second and third codon position. Support values are generated using a 100 rounds of the "Rapid" bootstrapping option of RaxML. The resulting tree. Well you can just say you downloaded it and viewed in fig tree to generate a publication quality image. With just a little tweaking. That is your materials and methods section under phylogenetic trees and pardon me while I do a rapid scroll down so you can see where all those references are, they're the references you need, but of course the main references, I used the codon tree pipeline at PATRIC inside our most recent publication. The materials and methods are taken care of. Let's go back to our first tree job. Here we are on that first tab. Then you say, the tree was built on a 100 single copy genes, this is in the results section, that included 53,880 amino acids and 100,683 nucleotides. That should satisfy any reviewer. Reviewers are used to seeing trees based on maybe the MLST genes, or maybe a few genes, these are a 100 genes, and that these are a number of characters. That is going to make any reviewer, if they have any qualms, they'll be very happy with the that, I guarantee it. That's where you find the information to report problems, if you have them. You can go to the tutorial and look at things if you're having issues with it. From the tutorial, you can grab text for your methods and materials section and also from the job report, this is for the results section. What's more, all of these tables are available in those downloadable files that I've shown you earlier. You can just give those a supplementary files to submit that with your paper, especially these. If a reviewer says, ''Oh yeah? What genes did you use?'' Well, here they are, here's a 100 of them. You can just download it as Excel and submit it to the reviewer as supplementary file. What could be better than that? This ends our code on tree instructional videos. Let us know if you have any questions or concerns. We really want you to be happy with it and we appreciate your interest and your use of PATRIC. Thanks so much. Bye. We're down to our final assignment. Usually I end these things with having you ask a question and report a failed job, I think you already know how to do this. For this assignment, I want you to go into those 32 jobs, and I want all of us to look at those things, and looking at the tree analysis statistics. Let me show you the best way to do that. This is one of the tree jobs. I would click ''View''. Then for the HTML, I would click ''View''. Look at that lovely lovely lovely tree. These are the statistics I want. We don't have TSV file for this so I'm just going to copy it this way. Then I've got my spreadsheet here, this is the spreadsheet I gave you, this is the one I've been filling out. Generally what I do is, I click on a file and I say it we'll do it as text. If I do match destination, sometimes it creates it all on the same line and I get very frustrated. I don't want you to get frustrated. I copy that, and then I go in and paste, and I transpose it. That's the way that I take that data in. Some of these jobs are going to take a long time to run so this is a long assignment, just for your own benefit. Some of the things I want you to think about when you look at all of that is, which seems to work better when you've got the 500 genes? What got me closer to including all 500 genes? Which combination institute that faster? On average, which took longer when you had more deletions, more duplications, or the combinations? Which were the longest tree jobs? You'll also know how long they took to complete. Now, unfortunately, the tree jobs record is running in seconds in that analysis statistics, but I think you can divide by 60 for minutes or even another 60 for hours. I think you can do it. Another thing I want you to do is look at the trees in the reports. At what point do you start to see the groupings solidify? It doesn't matter how many genes or additional genes, whether they're with families with deletions, their families with duplications, the tree structure stays the same and the support values stay the same. That way you'll know how much oomph you'll have to give for this analysis. Do all that stuff, and then if you have any problems, let me know. If you have questions on things, you go into the help tab and you get a hold of me there through ''Provide Feedback''. If you had jobs that failed, go ahead and do the report issue icon. You could see from my spreadsheet that I had a few jobs that failed. I'm interested to see if we all share the same kind of job failures, or maybe all of yours will complete, and maybe it was just that PATRIC was mad at me that day. This ends codon trees. Next we go into a detailed analysis of the protein family sorter. Another one of my favorite tools in PATRIC. I'll see you all there. Bye bye.
