 We can, thank you. Okay, thank you for the introduction. My name is Davide Kiko, and I am a researcher at the University of Toronto in Canada. And I'm gonna talk about this package called the gene expression from GO, which is another package that makes it very easy to read data from gene expression omnibus GO that is the world repository of gene expression data. Thank you, Hermione, it's in for a minute chance to present. This is the paper about this package that I published a few years ago. The package is available on CRAN. And so as many of you probably know, gene expression omnibus is an online repository of thousands of datasets of micro-region expression and of RNA methylation developed and maintained by the National Centre for Biotechnology Information, CBI. And the datasets available on GO can be used for new scientific analysis, or also for meta-analysis by selecting datasets that derived from the same gene expression platform. For example, if one wants to perform an analysis on data of cancer or multiple datasets, they can select datasets having the word derived from the platform affymetrics HGU133A, that is one of the most popular platforms of affymetrics. And by selecting datasets of the same, deriving from the same platform, they can be sure that they're gonna have the same probe sets, the same will be compatible between each other. And so they will be able to first of course do the batch correction and then do the scientific analysis on all of them. It may be fine to check if the genes, most relevant genes for a specific outcome found on one particular dataset are also relevant in all the dataset. For example, prognostic genes or diagnostic genes. And so this aspect is extremely useful for the generalizability of centric discoveries because being datasets of the same platform compatible between each other, they can serve as validation cohorts or discovery cohorts for the studies, which unfortunately this is not the case of a total care of records because as we know many times, most of the times datasets of the same disease, for example, but arriving from different hospitals, they have different features with different meanings, so they cannot be compared and say. The data on GEO come with the patients or the samples in the columns and the prop sets of the prop set of the platform on the rows where prop sets are sets of fragments of DNA known as hybridization probes and represent a precise point in the genome for that platform. So which means that it's exactly that specific point in the platform that represent a contained data data. And each prop set can correspond to a single gene, but a single gene can correspond to multiple prop sets, to two prop sets, let's say. So it's not a one-one relationship. So for example, the prop set 203072 underscore AT is a prop set for the GPL96 Affymaticx platform and it refers to the MY01E gene symbol that is a symbol of the, it's the myosin IE protein-crowding genes gene. And of course, as you can understand, it's necessary to perform a mapping between these prop sets and these gene symbols. And there are already some packages that perform that like GZ and annotate, but the usage of these packages can be a bit difficult for beginners. So on crowning by conductor, there are some software libraries that automatically download the data from Geo, in particular GeoQuery that is quite used and quite famous. And as I mentioned, there are other libraries such as annotate and jet set that can annotate the gene symbols to prop sets. But as I mentioned, they are a bit difficult to use for beginners because they have many fields and especially annotate many fields and the object that comes out from GeoQuery has a complex structure, a list of fields that might not be easy to use and understand by beginners and students. So a few years ago, when I was facing this problem, so I decided to develop a new R package that would do both these jobs in an easier way. And I called it the gene expression from Geo. And so it has two main assets. It downloads users to download a dataset from Geo in an easy way, just specifying the code parameter of the Geo dataset and it saves it into a data frame. And the second feature that it has is that it associates the gene symbols with the prop set of the downloaded dataset. And this feature is, this asset is available for 19 different platforms, not for all opportunity. And yeah, you can install it from crown, install the packages gene expression from Geo, load it with the usual command that you all know, library gene expression from Geo. And let's see an example. Let's pick this dataset called GSE 3268. It's a dataset of gene expression, profiles of squamous lung cancer cells used to identify genes that are different, differentially regulated. That's the title of this dataset that can be found online on Geo and was released in 2005. The comments to use gene expression from Geo are very simple. So first we have a flag as it gets symbols to genes that we put to true because we want to retrieve the gene symbols of the prop sets and they were both parameters that are true that says that it can, we want to print the messages and dataset code that is just the code or dataset that we want to download GSE 3268 that is what we just saw in that example on lung cancer. The call is a very simple gene expression from Geo, dataset code as it gets symbols to genes were both online. And so this comment will download the dataset and associate the genes symbols and we'll save the data into gene expression, the F as they call the variable here which is another frame of approximately 22,000 rows which are the prop set and 11 columns that are 10 samples of the dataset and the gene symbol variable. So to adjust the screenshot of the execution of this comment, the dataset, if you put verbose equal to true, it will print out the information like the features contained in the dataset and then it will start a loop to associate the gene symbols, the prop set ideas. The results, it's a feel like the piece that you can see in this slide which I printed with a head command. So on the rows there are the prop sets, the original prop sets of the platform for this dataset. So 1007 underscore S underscore 80, 1053 underscore 80 and so on. On the columns, you can see the samples except the last one that is a new column called gene symbol that associated the gene symbols that it found to the corresponding genes and for the corresponding prop sets. So for example, the prop set 1053 underscore add was associated to the gene R of C2 and the following prop set to the gene HSP86 and so on. For the last gene, for the last prop set 1294 underscore 80, the gene symbols found were two and it's also possible that some genes might not be found, some gene symbols might not be found. So the first row that you see, didn't find the genes, the gene symbol for the prop set 1007 underscore S underscore 80. For the- Just a few, less than a minute left now, just FYI. Yeah, and then for the future developments, I plan to insert to allow users to use annotate more platforms and possibly I would like to add the parameter for possibly automated batch correction. Is this the end? And yes, again, who wants to know more about it can read the paper or can write me an email and I will be happy to answer any questions if you're out. Thank you. Davy, thank you so much. That was a really wonderful and very interesting working with these large repositories of gene data can be quite difficult and these tools are incredibly useful. Thank you. Are there any questions for the audience? All right, well, if there are no questions at the moment we will, we can continue those in the chat. If you-