 I am Jorge Piñales from the University of Minnesota, and I'm going to introduce to you a new science gateway that we are developing built on top of Galaxy to assist a new domain science, the X-ray microscopy community. The X-ray imaging of microstructures gateway is what I'm going to be discussing this talk. It's a collaborative effort involving the Science Gateway Institute at the San Diego Supercomputing Center where Mark Miller and Marius Huang are the developers. The Cornell High Energy Synchrotron Source, or CHES, which is where the synchrotron facility is located, that involves Devin, Buji, Matt Miller, Kelly Nigren, and Werner-san. Bob Souter is also working with us from Carnegie Mellon in the Physics Department. And finally, at Minnesota, both Jim Johnson and myself are participating in the effort, which is funded by the National Science Foundation, the Office of Advanced Cyber Infrastructure in collaboration with the Division of Materials Research. X-ray diffraction imaging is used in a variety of fields. We are focusing here in material science specifically. The experiment very schematically involves shining and coherent X-ray beam on a sample and then collecting on a screen the diffracted spots that are produced by the sample. The sample itself is being rotated and the detector location is also being moved between the far field and the near field so that in the end, a three-dimensional dataset with diffraction spots is produced for every experiment. This is a large dataset that contains an intensity map of the scattered radiation. This set contains material-specific information that has to be deconvolved and it includes not only the crystalline structure of the material, but also because it's really not a single crystal, it's microstructure, it's decomposition into grains in which each grain is a subset of the material exhibiting a different orientation. The drivers for galaxy adoption in this specific field are fairly common to others. There is a barrier to users to learn the infrastructure and to deploy their own tools. The synchrotron facilities are fairly complicated and the computational facilities around the synchrotron are also fairly complex. Currently, chess personnel has to instruct every user, which is of course not very scalable. Synchrotron data is very large and therefore hard to move. For each experiment, a user would collect a few terabytes so there's a need to manage and reduce the data before it's moved for analysis. Collaboration is difficult because data owners have large data sets which are hard to share and they're difficult at the moment to make public, especially given access to the modeling community which doesn't really have access to the experiments. And of course, users want to have access to public data sets to benchmark and try tools that they may be developing. With the gateway, we hope to facilitate the analysis and manipulation of the data from everywhere, access, discoverability of data sets that you would not be able to find otherwise, share data and workflows, collaborate with other users but also with modelers that typically do not access the facilities and access and integrate new tools that are being created by the community continuously. The users therefore would not have to learn the details of the synchrotron file systems. The data will be presented uniformly. They will not have to install their individual codes. The latest will be available automatically. They will not have to learn how to interface with the local computing systems. The benchmark of the codes will be provided to make sure that they're efficient and they will be run either locally or at X seat at the connection that we already have depending on the code. Workflows as they are typically very complicated. They start with raw data currently of about three and a half terabytes per experiment. This is complex to show you the purpose. Input is required in terms of the material that's being examined, in terms of the reconstruction and reduction that's going to be used and on the output that's being produced. And that has a wide variability from experiment to experiment as different modalities can be used. Of course, Galaxy extracts these or abstracts these in ways that you know. A few challenges that we're addressing at the moment that might interest some of you are a fair, large reliance on interactive tools. So at the moment we have built notebooks into the Galaxy server and also we have access to ParaView, a visualization framework. We are also building connections to modellers including mesh constructions, data desolation. So this is active to people. This is useful to people that work on plasticity and need to investigate for example, the distribution of stresses in a polycrystalline material and also ways to standardize file submission forms for people that are using this already. We are contributing what we do in terms of community resources. There's a new category, the Galaxy toolshed which is called Galaxy structural materials. That's already there, not many tools available of course. There is a development repository in GitHub which you can access. And of course last is the server address which is also running for those that might be interested in accessing it. Thank you for your attention and I suppose I'll get some questions in some modality. Thank you.