 first of all came out of a project called the Neuroscience Information Framework and the Neuroscience Information Framework what basically did is it asked for us to put together all of the important information and the data from the different databases of anything that anyone would want to know about neuroscience. So we took on this task we were at the lead institution at UCSD out of a five institution project. You know some of the other institutions were fairly famous ones as well. Harvard was one of them Yale is another Caltech et cetera. And so what we were able to do is we were able to start putting together a system that would allow us to put a lot of data together. But one of the things that we were asked to do by the National Institutes of Health was to actually help researchers find antibodies because it turned out that lots of people were looking for antibodies and this was kind of a waste of time. I mean why wouldn't people want to be able to just figure out which papers are using which antibody. So we were asked to do this as a way to essentially automate this process of antibody finding. Well we while we were able to do many things we completely failed at this task and we completely failed at the task because the task was too hard. The information wasn't in the papers too for us to find. So there was nothing that we could do with our algorithms or any of the fancy math that we could possibly bring to bear on this problem to actually put into the paper something that wasn't in the paper to begin with. So that's what actually started us trying to think about what are the ways that we could actually get journals to include additional data. So that would help actually not only us to you know do this this text mining task but also help other readers of those papers to find the antibodies and that grew into all other kinds of research resources. So the mice, the flies, the cell lines and of course antibodies and cell fur tools. Those are all the things that people were looking for that they couldn't find. They couldn't readily identify but if we can if we thought well if we can get people to start to put in a little bit more of this information into the paper straight away then we would be well the task would be much easier for us but it would also be much easier for the readers. So how it got started is essentially again it was us failing to be able to find these antibodies and then the second piece was well how do we address this problem and the way that we address the problem is we brought together a lot of the editors into a room and they said well here is a possible solution but it's going to depend on having these indexes so what an index is is a whole list of numbers. So you can imagine this like your GenBank identifiers. In order to publish a genetic study you need to find your GenBank identifier. If you want to make sure that you're talking about the same SNP you need to get an RS number for it. So this is the same kind of thing we just have a way of identifying mice, we have a way of identifying flies, we have a way of identifying antibodies and cell lines and when we started this one of the things that the authors or the sorry the journal said is we can't send the author to 20 different databases and we're going to need to do that. So what we would really need to do is bring all of these different databases into one place. We have one uniform way of accessing all of this different data and one uniform display. So with our experience in the neuroscience information framework we were able to very quickly create a portal that would be able to actually do this and allow people to search across all this different kind of data and when they're able to do that they're usually able to complete the task fairly quickly.