You can add location information to your Tweets, such as your city or precise location, from the web and via third-party applications. You always have the option to delete your Tweet location history. Learn more
To understand my skepticism of some microbiome research, first you have to think of the outcomes that are being studied... 1/
The microbiome, being made up from a large number of different organisms, is of course highly multi-dimensional, even at the phylum level. One (of several) challenges with this is the large number of "researcher degrees of freedom" these kind of data afford us.
I could look at overall measures of diversity (and there are several of these), as well as changes in specific taxa, of which there will be many.
So for some studies, I would be much more impressed if the researchers told me what they expected to find and why before they analyzed the data.
Most of the time, simply saying that changes were found is profoundly boring in my opinion, because we now know how sensitive the microbiome is to many seemingly benign factors. https://www.nature.com/articles/s41598-018-21986-7 …pic.twitter.com/yG8aUOAp4V
Another problem is that microbiome data are compositional in nature. Due to how the microbiome is measured, what we usually have are proportional measures of the different organisms.
Thus, if something goes up, then something else (or many somethings else) must go down, since the total proportion must still equal one.
This throws off "traditional" statistical methods in ways that many still don't seem to understand, despite this being a problem that goes back to Karl Pearson. https://www.frontiersin.org/articles/10.3389/fmicb.2017.02224/full …pic.twitter.com/25c3vehlNO
Recognizing this problem is just half the battle though - let's just say that actually dealing with it is on the "advanced" side of stats and maths, and imo relatively few people out their appear to be credible in their approaches.
So if I think about the general lack of statistical thinking and input into many "simple" analyses, I find it hard to believe that even most of the groups out there analyzing microbiome data have got a grip on it.
The final thing, which is way out of my wheelhouse, is that generating the data for analysis is a really tricky thing in and of itself, requiring a high level of expertise. So beware! https://www.nature.com/articles/s41579-018-0029-9 … /end
Well, this is awesome. Say more stuff all the ttime *if you want.)
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.
