Science synopsis
Dec. 28th, 2016 05:56 pm![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
rebeccmeister[As posted to other social media]:
As you undoubtedly know by now, one of my favorite George Pocock quotations is: "No one will ask you how long it took to build; they will only ask who built it."
Nonetheless, at the end of a long day of data quality-control and analysis I find it satisfying to tally up some numbers.
So far, this circadian experiment has involved gearing up to run time points 45 times, across a 5-month period. I've poked and dissected 393 crickets, and we'll wind up using data from 211 of them. In some respects this is a far cry from the 6 different radiotracer experiments from 2015 (averaged ~250 crickets/experiment), but of course there are some major qualitative differences between experiment types. Overall I would say the radiotracer experiments were less grueling.
And regardless, that's a lot of crickets to have poked. Thank you, crickets.
-
Also, I suspect I'm not quite finished with it all. I'll take my current analyses and will run with them for now because I'm giving a conference presentation in one and a half weeks, but I am thinking that I'll have more confidence in these results if I boost all my sample sizes to closer to 9 crickets per morph per tracer per timepoint (so 9 crickets x 2 morphs x 3 tracers x 5 timepoints = 270 total). The good news is that if I decide to proceed, I at least have my cricket-rearing protocol in good shape, so completion should be straightforward. In addition, I'm feeling fairly confident about my method for characterizing lipid oxidation.
It has been quite a ride.
As you undoubtedly know by now, one of my favorite George Pocock quotations is: "No one will ask you how long it took to build; they will only ask who built it."
Nonetheless, at the end of a long day of data quality-control and analysis I find it satisfying to tally up some numbers.
So far, this circadian experiment has involved gearing up to run time points 45 times, across a 5-month period. I've poked and dissected 393 crickets, and we'll wind up using data from 211 of them. In some respects this is a far cry from the 6 different radiotracer experiments from 2015 (averaged ~250 crickets/experiment), but of course there are some major qualitative differences between experiment types. Overall I would say the radiotracer experiments were less grueling.
And regardless, that's a lot of crickets to have poked. Thank you, crickets.
-
Also, I suspect I'm not quite finished with it all. I'll take my current analyses and will run with them for now because I'm giving a conference presentation in one and a half weeks, but I am thinking that I'll have more confidence in these results if I boost all my sample sizes to closer to 9 crickets per morph per tracer per timepoint (so 9 crickets x 2 morphs x 3 tracers x 5 timepoints = 270 total). The good news is that if I decide to proceed, I at least have my cricket-rearing protocol in good shape, so completion should be straightforward. In addition, I'm feeling fairly confident about my method for characterizing lipid oxidation.
It has been quite a ride.
