[week] always fun when you have to break out the pencil and paper to figure out where am I what am I doing

taking a linear set of coordinates and converting them into circular coordinates zeroed on an arbitrary point is a little brain-enfuckening for me

oh NEAT enfuckening but daaaaaaaaaaaaaamn

StupidSexyV⚠der : I have very little formal math training but fortunately I have an okay ability to visualize problems. The challenge comes in converting that intuition into code that my computer can understand.

/nods

with my learning disabilities math is something that is a VERY REAL struggle for me but it's seeing other people do it and I'm like WITCHCRAFT!!!!!!

....and I still calculated some of the scaling factors wrong but with more diagrams and an excel spreadsheet, I now have half of the calculations done!!

yeah. fuck circular coordinates. unless it's capacitance. then it can stay

seeing imaginary numbers handled as circular onto the linear number line is enlightening

and calculating capacitance is the only physical application I know

squints
are you trying to re-derive cartesian to polar?

because, not to denigrate your work, if that's what you're doing, There's A Formula For That

also, polar coordinate systems are very useful

useful, yes

but they can heck off most of the time

look my degree is aerospace and i used to do mission design
polar is everything

yes that is a case where they are best

6LilacMimeLions : Unfortunately not, it's DNA coordinates (basically a number line) that have to be scaled to represent their position in a loop structure, relative to features within the loop.

mapping

My thesis project centers around interactions between two classes of regulatory features that interact to affect how genes are used, and the scope of these regulators turns out to be broadly defined by whether they're located in a looping structure called a Topologically Associating Domain.

yep

I'm generating a plot to prove that using this fucky circular TAD mapping is actually more efficient for determining likelihood of regulatory action than the traditional linear coordinate system.

there's a lot of domain mapping in computational fluid dynamics and that's the class i got a gentleman's c in

and by "prove" I mean "look the percentages are happier"

XD

the percentages are happier

a happy little percentage

I totally whiffed my calculations in like five different ways today before I finally just threw an example TAD's features into an excel spreadsheet and figured out what scaling factors needed to be applied based on relative position to my reference points and the TAD boundaries

...and then I still got nonsensical results for a good half an hour until I realized "oh right"
"don't take a logically-defined subset off of the data that you're actively changing when you have to do it in sequence"

w h o o p s

o o p s