So I just take this analogy from [UNKNOWN] slides sometimes when I talk about asthma. So if you think of different kinds of asthma, allergic sensitization, and the path of genesis. You know you think you might have some allergy exposure, which you brings into your body, the specific [UNKNOWN] you know [INAUDIBLE] which has been sensitized and these [UNKNOWN] that to our mediators. And these plans where mediators create classical symptoms, of bronchial construction, mucus, generations like that. And the way that we can attack this, is primarily through medication So you have like your inhaled corticosteroids over here. You've got your anti-leukotriene type modifiers over here. You've got your anti-IgE type stuff over here. And, and so that all does varying type of jobs of, of, affecting people's asthma But then, there's also this part up here. So one of, one of the reasons I'm excited about a lot of this environmental intervention stuff that we're doing is that it kind of affects the most upstream part of the disease process. And not only does it affect this most upstream part. But it's actually very inexpensive relative to all this medication. So, it's efficient, and it's potentially powerful. And so that's kind of this, that's kind of nice. So if you were to translate this over, to kind of a way that research is disseminated so there's people, people conducting research over here that may be problematic, it may be incorrect in terms of the analysis. They submit a paper to some journal, that goes through some, the, and the paper gets published, and after the paper's published, you get the data and you get the code. And you can see what they did. So, the typical kind of safe guards are, [INAUDIBLE] maybe editor thinks it stinks so he rejects it. Or if they send it out then there's peer reviews. And the peer review's supposed to be a check on something. And if it, then if it gets past your review it gets published, so that's where reproducible research comes in at the most downstream aspect of the process. And so there's a lot of discussion now about doing this, and so the idea is that okay well why would wait for all the data and the code to available after the thing is published. Why not do it as part of the publication process? And I think that's, that's a noble goal, but I think totally unrealistic. Because I don't think it's realistic to expect peer reviewers to kind of reproduce everything. I speak from experience. So Of those of you who are not in statistics, you know, are not used to waiting the typical six to twelve months for your review to come in. And if you add on top of that the need to reproduce the entire analysis you would be promoted 40 years from now. So so, so that's reasonable and I'll give an example of that. But I think the questions well, is there something we can do? I mean in the most upstream part of this process. And so that's what I want to talk about today. So this one example of moving the reproducible part to the peer reviewed passport, it's not exactly up here but it's somewhere up screen a little bit. This is just, we have a policy at the journal of bio statistics, which is easily the best statistics journal in the world, and and I'm the associate editor for what's called the associate editor for reproducibility. And we have a very light policy, it's totally voluntary, where you can submit your date and your code to the editor, the associate editor for reproducibility and he, [INAUDIBLE], will reproduce it. And if you if it's not a very, it's not a hundred percent fool proof policy but it's never happened that I haven't been able to reproduce anything. But if you, it is really useful, you get this little R stamped on the top of your paper. Now there's two other options, there's a D and a C. If you submit your code, then you get a little C over there. I don't check that, I just assume that's fine. If you submit your data, you get a D. If you submit your data and your code, you get a D and a C. And if I reproduce it, you get an R. So and then this is Jeff's only contribution to this work. [LAUGH] And [UNKNOWN] these papers. He got an R. And it took me a week to reproduce it, but it was fine. So anyway, so that's one model. This is not really a kind of a deterrents, so to speak, because it's old voluntary happens after the paper is deceptive from pure evil. So but, anyway so this is one model [UNKNOWN].
