 All right, so Dr. Morrison also comes to us from the Van Andel Institute from the Schoen Lab, and Jacob will be presenting bisplotty today for a graphical representation of epi alleles in epistates. Oh dear. How do I share this? It's not on there. Go ahead. There we go. All right. Good morning, everybody. As Tim mentioned, my name is Jacob Morrison. I'm a bioinformatics research scientist in the Schoen Lab at Van Andel Institute, and I will be talking about bisplotty today, a tool to be able to create DNA methylation analysis plots. So, just as a very brief overview for those of you who don't know what DNA methylation sequencing is, this can either be bisulfite-based or enzyme-based, and what happens is a unmethalated cytosine will be converted into a uracil during this process, whereas a methylated cytosine will be left as a C. The uracils are then turned into thymines during PCR amplification, and what happens here is that the epigenetic difference that you start with is then turned into a genetic difference, which you can probe using sequencing and then be able to tell during the alignment stage where your methylated and unmethalated cytosines are occurring. So where does bisplotty fit in here? So, first, bisplotty is an in-development R-based package for plotting DNA methylation analysis figures using standards-compliant file formats, and where this happens is you do your sequencing. I guess there's no laser. So you do your sequencing, you do your alignment, you extract your methylation, do your analysis, and then you would use bisplotty then to create figures that you can use to then analyze your analysis that you've done or to be able to put into your papers and presentations. So the rest of the talk will be just going over the different types of plots that exist already and then what we plan to do in the future. So just to start, we have just Kennedy's standard global methylation plots where you can look at the density of methylation values across samples, whether it's 1D, which is not shown here, or 2D, which is shown, where you can compare two samples on a per-CPG basis. This works with inputs from BSEEC objects, so those of you who have used BSEEC before, as well as those who do like EPIC arrays or 450Ks, you can use beta values from these matrices in this as well. The next type of plot that we have are what we call multiscale plots, and these are based on Kneidenberg et al. in Nature Methods of 2014. And what this does is it shows the average methylation values across many bin widths. So you can start with small bins and then compare where you have short-range methylation interactions to long-range methylation, and so you can get this in a single view here using these plots. So these inputs are generated via snake pipe line, which we have available on GitHub down at the bottom. Also have it in the summary as well, and so you're able to do comparisons across samples this way, as is shown here in this example of a newborn set of cells in a 103-year-old set of cells. And then we also have read and fragment level methylation plots. And so these plots use what we call the epi-bed format, which is a compact information-rich bed-compliant format file that we have developed at VAI in the Schenlab. And what these plots do is each row here that you see is an individual read, or in this case specifically a read fragment where for paired-end sequencing you can combine the information that is correlated in that paired set of reads and then you can look at that as an individual entity. And then each column here then shows a single CPG in the genome, or a SNP as is seen here. And then you can also plot the average methylation, so it will average each column into the average value of the CPG methylation that you see there. And then by being able to visualize SNPs, you can be able to see allele-specific methylation or epistates as well. And so another way that you can look at this is to look at long reads. So the previous example was short reads. Another way is to use long reads. Currently, I want to mention that this only works for side-of-zine converted long reads, so in this case EMC plus pack bio. But our goal is to add the mod-bam functionality in the future, which has been a recently developed availability from SAM tools in the SAM specification. Also, by being able to look at read or fragment level data from a bulk sample, you can get a quasi-single-cell visualization out of this because a single read will likely come from a single cell. And so you can be able to tell differences that way. And then finally, with this type of plot, you can also look at NomeSeq data. So you can be able to visualize both methylation from CPG methylation, as well as cropping to an accessibility via GPC methylation. And I want to mention with these types of plots that both read orders in the CPG plot and the GPC plot are ordered the same, so that you can be able to tell a combined between the two. And so finally, just some future directions and summary. Right now, it is only available on GitHub, but we are planning to aim for the October release this year, 316 of Bioconductor. And we're aiming to improve the existing plots I showed today, including multi-sample plots, adding annotation tracks and other things. We're also planning to add additional plotting. Obviously, there wasn't money that we showed today, but we're looking at plots that show the methylation that surround different genomic features, CTCF sites, promoters, enhancers, so on and so forth. Differentially, methylated regions is a very common analysis, so we want to add plots for that. Looking at average methylation values in chromate GMM states, as well as if you have any plots that you tend to use when you do DNA methylation analysis, we'd love to hear those. We'd love to look at getting those added into Bisplotty, so feel free to reach out to me or any of the other authors on that. Finally, Bisplotty, like I mentioned, is an in-development, our base package for plotting DNA methylation analysis figures using standards-compliant file formats. We already have several plots types that are already implemented, and we're looking to add more in the future. So here are GitHub links for those, and as I mentioned, Bisplotty will be available hopefully in October on Bioconductor. I just want to thank the other authors, Ben, James, Ian, Wei, as well as other members in the Shinn lab, Wandaing and the Joe lab, and then obviously funding, so thank you very much. Thank you, Dr. Morrison. Thank you to all of our speakers in this section. Drs. Ben Gurie, Sato, this is Harmon, and Dr. Morrison. If we have questions in person, please feel free to use one of their microphones. If you'd prefer to ask a question in chat on Webex, please feel free to do so. We have about 13 minutes till the next round of sessions, so if we don't have any questions, you might also feel free to avail yourself of a refreshment of some sort. Thanks everybody for attending here and online, and I'll see you in the next section. We'll take a short break, and we'll keep watching the chat for any questions you might have. See you in about 15 minutes. Thanks everyone.