 Okay, thanks. My name is Travis Wheeler. I'm an associate professor primarily housed in the department of pharmacy practice and science, but with affiliations across. A number of other departments that that begin to sound an awful lot like somebody who thinks about computing computer science data science and school of information. And also affiliated with the genetics program and several others. I'm going to follow a format that's similar to the one that that Bonnie just followed in terms of describing who I am what my research interests are and and opportunities for collaboration. I'm going to keep it at a pretty. High level. The one thing I want to say is I'm sorry that I can't be there in person, but you're welcome for not being there in person. I am very sick and it's amazing that I can talk right now. I run a pretty large research group. These are folks in the group, some of whom affiliate with things that I'll talk about going forward. In addition to running my own research group. I'm the director of the University of Arizona health sciences bioinformatics group, which is a group of informaticians. On the the U of a health sciences campus with a capacity building mission. Right. So that means that they collaborate on projects across campus in particular and that within health sciences that are in need of bioinformatics expertise. These are people who who have expertise in bringing computers to bear on analysis projects, which should sound like something that we all care about. And they aid in analysis of existing datasets help develop plans for new analyses. And in general serve in that capacity building mission. I'm also associated with the all of us research program. Many of you may be familiar with that it's it's in some way analogous to it provides data that's in some way analogous to the data that that's available within the VA system. Perhaps on a smaller scale in some ways, but a wider scale and others involves genomic data and electronic health records for currently a little under half million people, but with the numbers scaling up with the target of a million with a with a focus on on population diversity. I think diversity. And my team provides an informatic support on among other things on seed grant projects. And research interests. I would describe myself and my group and in general as a collection of generalist computational biologists. We develop algorithms and machine learning or artificial intelligence, fundamental approaches and engage in serious software engineering and if you, I guess my video in the corner shows we also build some visualization tools along the way. A lot of that focuses on genomic analysis we've got a collection of tools that are really heavily used in the realm of analyzing or annotating what's in genomes or microbiomes the bacterial collections in in a variety of environments and I won't talk about any of the details of those but there are a lot of algorithms and machine learning approaches that that are at play there. We're also involved in projects related to the field of drug discovery. So you've got a protein you'd like to identify a drug that will work effectively with it and we built a tool system and infrastructure that can scan collections of billions of candidate molecules to identify promising candidates. And we're in the midst of building a petabyte scale pens or hundreds of petabyte scale collection repository for Adam level simulations of proteins and the drugs that they bind called MD repo MD repository. And on the side, I don't know if this will show up in anybody's field of interest but we engage in in animal tracking and behavior classification and recorded videos, which, although it sounds a lot different from the other two things ends up boiling down to the same kind of context of a modeling and algorithm development. We have begun relatively recently getting involved in this area of health outcomes or precision medicine in the context of association studies so I've got a generic picture of a genome study that identifies particular regions along a genome that are appear to be associated with a particular phenotype. And those are pretty effective at identifying individual positions but it may not find applied traffic effects the combined effects that that may inform about either health risks or or potential beneficial outcomes and in particular may not effectively blend both genomic and healthcare record data, and we're building machine learning approaches to be able to to extract large more complex networks that are otherwise very difficult to to call out of the data. The open to collaboration slide so I'm not going to name specific things. I think in this kind of a talk is a flirtation of sorts I'm trying to convince you in this speed date that you want to talk to me. I'm a generalist and I love talking about research opportunities, but what kind of research opportunities matters right so in particular, we are useful at and interested in cases where there are current analysis methods. There's software that does some kind of analysis that's either inaccurate it doesn't really work as well as you wish it did, or it's too slow it can't handle the kind of the scale of data that you've got it works for 1000 patients but doesn't work for a million. Algorithms and statistical models to be able to extract data out of extract signal out of data and to do it very quickly. And often our best utility is when we are taking methods that exist already and they just need to be improved a lot. So if you've got that kind of a scenario we're a great group to be talking to. Also identifying if you've got existing tools that that do a thing pretty well but they're missing out on some particular offshoot of the analysis. And I'm especially interested in in documenting and understanding the impact of uncertainty in labeling or classification or identification of things I think uncertainty. A lack of knowledge of a prediction of the confidence of analysis is really missing in a lot of analysis and it's super important to to predicting how things are going to turn out we work well with huge data sets. And this is the last collection of bullet points is what we do we build algorithms we build AI or machine learning methods. All in open source software will end as I mentioned in the previous slide we build complex data repositories these sort of petabyte scale repositories I've got plenty of familiarity with as well and so between building and interacting with them in the context of all of us data. I am aware of the issues that arise in the kind of data set that data sets that exist at VA and I'm pretty excited about finding the right niche to be able to work with you to improve analysis of those data. Thanks.