kstreet13
commented
2 years ago Hi @CodeInTheSkies ,
This is a tradeSeq function, but briefly, the values represent the predicted mean smoother, for which we use smoothing splines (not a moving window, which would be more like a loess smoother).
And I'm not entirely sure what you mean by doing it manually on a section of a matrix (I think of the basic tradeSeq inputs as a counts matrix, a vector/matrix of pseudotime values, and a vector/matrix of weights), but feel free to check out the code, here.
Best, Kelly
Hi Kelly,
I'm trying to understand a bit more deeply about the smoothed heatmaps such as the one here: https://github.com/kstreet13/slingshot/issues/142#issuecomment-885709765
How exactly is the smoothing done? Is it a simple moving-window average? I'm wondering, if it's not too complicated, is it simple enough to manually do on a section of a matrix, for example? If you can point me to the code for the smoothing function, that will be great!
Would appreciate your input whenever you get a chance.
Thank you!