So now that we have created the annotation files, as well as the alignment files that can be viewed and curated in the integrated genomics viewer. We are ready to load them into the viewer. But first, a couple of words about where you can find this tool. You can go and Google integrated genomics viewer, and you can either run it from the Broad Institute's website, or we can download a copy. So downloads here, and you can have your own copy after registration. So now that you downloaded your own copy, all you have to do is to click on the icon. This will open IGV. And then, we're going to load into IGV, our data sets. And let's start with the transcripts. Let's start with a test. So first of all we want the human genome H19. That's the one that we used. But if we're wanting another genome, we could find it here in the list. Or we could even load a genome from a FASTA genome file that we could provide. But let's start with human genome, HG19. And now we're going to load, let's start with the annotations. So in this case, Desktop > Coursera > IGV > GTFs and start with Test1.sorted. Okay, you also have a listing, these are the RefSeq reference gene annotations, so we've loaded Test1. Load from File, Test2. We won't see anything until we locate and we narrow down the region and zoom down to see the features. And now let's also load the controls. And just for the purposes of illustration, let's go to our gene TMEM215. So now as you can see in the viewer, we have one panel. And we have stacked panels with each panel corresponding to one of the annotation sets that we provided. And we can view these annotation in the Collapsed form that is projected along the genomic axis and collapsed, Expanded and Squished. I usually like to see Squished so they are extended but not too much so. And we squish the alignments for each of these. One can also change the track color. So for instance, in this particular case, let's make this red, and so on. And now you can see that the pattern that emerges is that of a gene with two exons. And there's an additional single exon transcript that might be a transcript, or might be in 24 other type of noise. As are these. But what you will observe is that the top three panels, corresponding to the test examples, have a clear definition of the gene. Whereas the bottom three, where only one of them has enough reads to allow reconstruction of the gene. So the other two barely show any signal. And you might recall that the average gene expression level was 0.18 compared to about 6 for the test examples. So now let's look at the alignments to verify that this indeed is the case. So we're going to Load from File now, let's load an alignment file. We're going to go to IGV > BAMs. Let's load the alignments for the control, for one control, and we can load the alignments for one test file. And I might actually want to move them around to have the test there. Okay, followed by the controls. But for every alignment file, we have three sub tracks. We have one showing the coverage. So that's calculating at every position on the genome the number of reads that align at that position, and you can see the coverage varies. Then we have the junctions track, this shows the splice junctions coming from the splice reads. And some information about the depth, how many reads are supporting that. In this case, 37. And then we have a detailed view of the alignments in the BAM file. And here you see this horizontal line representing the spliced exon. So as you can tell there is significant coverage, the maximum number of reads is 139, I'm assuming at this point for the test file. Whereas if we're going to the control file, we barely have any reads and the coverage is up to ten reads. So, and if we load the remaining files, Test2, Test3, versus Control2 and Control3, we will see a similar example. So as you tell, we identify that this is a two exon gene. That is significantly, we can validate that is significantly more expressed in the test samples compared to the control samples. So this is one illustration on how we can perform a very simple analysis and curation, verification of results with a Tuxedo pipeline using the integrated genomics viewer, IGV.
