 No, this is not my talk. Okay, sorry, yeah. So good evening, everyone, and thank you for coming to my lightning talk, basically. And I have prepared more than 20 slides, but I tried to be quick and to beat the time restriction. So yeah, today, and I just was to share some actually recent work that developed in my lab. Instead of talking about the biological discovery, I will do more, I think, focus on the algorithm, the pipeline that we developed, which can process in-code data. So in early 2000, we know we have the first human genome draft that was completed by our public and private sequencing consultant here. But the time and efforts involved in the sequencing were tremendous, which is not realistic to process or deal with lots of individual samples. So that come to the next-general sequencing technology. From this figure, and also pull out from our NHGRI website, as we can see, the sequencing cost has been tremendously decreased. And so now it's more realistic to make multiple sequencing, like finish at a lower cost in a very short time period. So the in-code project, I think everyone here knows the purpose of the in-code project is to identify functional element. It has lots of sick data. So here, actually, I want to focus on the RNA-seq data, which basically is just bottom-of-side. So I just use this data to study the alternative splicing. So the in-code data set and from the one that I downloaded, actually mainly Tom's lab, so there are lots of cases that actually has not just a single end. It has a parent and also lots of replicated samples have been performed for the sequencing. So alternative splicing I mentioned earlier is like from one gene, you really could be transcribed into multiple MRs, right? And from an MRA level, and then could be translated into different ASIL forms. So this makes our gene expression more complicated. So corresponding, there are lots of tools that have been developed to study the alternative splices. So we have developed a tool called the Ritz-Blig walk-in initially, and now the current word is called the Ritz-Blig run. Basically, this is a software of the pipeline which can identify the novel splicing junctions at genome-wide level. We compared our tools to other some typical splice junction detection tool, like Tophead or Star and some other things. And our tools actually outperformed the identified special type of the splicing mechanism, both in the accuracy and the running the time things. And due to time restriction, I will not take the time to introduce this biological story behind for discover this algorithm. So instead, I was to talk a little bit about more the recent update about this pipeline, so called the Ritz-Blig fly. And this one actually has been improved a lot recently by increasing the sensitivity and the decreased running time. So specifically, the newer version can be, let's say, applied to rescue the read halves from the BOTAMX file, which is used in the pipeline as the first alignment step. So this increases sensitivity a lot. So if we look at the comparison from previous word, Ritz-Blig run with the newer word, Ritz-Blig fly, and we got the mouse data here, which is not encode data, but as a positive control when we develop this algorithm. So we got both the number of mapped reader for both the experimental conditions actually significantly increased. So running time in terms of the memory usage and the CPU time, we all get a great improvement. So what does this say is that this newer algorithm can process a large number of data in a quick manner and a more efficient way. So in addition, this new algorithm actually can extract the supply sequence after we detect the supply injections and to study its biological function. So the first focus we was to look at is what type of intro sequences that have to be extracted and what's the prevalence for the data set that we studied. So the two type of the intro, right, and we know that you really have the supply so why is it like U2 type? The other is U12 type. So U2 type more like the major splicing use different splice factors. U12 is a minor splicing. So they have different splicing signature for the like GTAG and U12 also has ATAC and not just GTAG cases. Their branch policy sequence also are different. So what we have done for this algorithm is that we have used our new algorithm to apply the encode data set that we compare the prevalence between U12 and U2 type. So interestingly we found actually there are lots of U12 signatures occurs in the novel supply sequences. That means that those are not followed the traditional canonical supplies of things. And this one also said and there could have lots of interesting biological signatures available in the encode data set. So the RSF pipeline also has a blast function which actually use the micro database to look at some micro RNA signatures in their supply sequence. So we actually also did a brief check for this micro array database to see whether there's a prevalence or there are some diseases associated micro is available in the encode data set. So this is also for the same 21 encode samples and we do see some micro arrays have been actually reported actually more than half of this 21 samples. That means there are certain micro arrays could also be prevalent. And also some of them actually two of them have been involved in some type of diseases. So in the summary and this said our algorithm can process the encode data and to identify some interesting biological signatures. And right now the preliminary data for this 21 samples and we have noticed that lots of U12 type non-connected supplies actually is available in this encode data set and more than the U2 type. So we have developed a website for this software and this is still the order word in the list to be wrong and they can take the user input files that report the output. And I think next working for ISR for ISR we would like to actually take the encode samples through the website that classified into the tissue type and Tom just give a talk today I think it'd be very, very cool maybe separate them into different cell types and then report the findings. So this is my lab and so Dr. Knie is on the left side and Dr. Gearsor is a Meritus Professor he's actively still involved in the research and the fund who is the person who developed this pipeline and now the work at Google. So that's pretty much today I want to introduce and thank you very much. Thanks. Yeah. Thank you.