Plotting the terminal half-life

[john-harrold]

I'm trying to generate an overlay of the terminal half-life for visualization. I can get the slope from the half-life. Is there a way to get the intercept from the PKNCA output? Or is there a better way to get at what I'm trying to do?

[billdenney]

I use tlast and clast.pred along with the slope to get the intercept. Then, you can use the first to the last point from the selected points for the range to plot.

[john-harrold]

I can just use the point slope for that.

Do you think it would make more sense to fit a line through the points used for the half-life calculation and show that over the range of those points? I can do it to the data for showing on a linear scale and for the log of the data on a log scale?

[billdenney]

Fitting the points should result in the same results as the half life points. I'm not sure that I understand the question.

[john-harrold]

I assumed that the tlast and clast.pred wouldn't be on this line but rather they woudl be used in the fitting? I assumed that to get the half-life all of the points from PKNCA::get_halflife_point() were fitted with lm() or something similar. So they wouldn't be on the line necessarily. What I want to do is show the actual line that shows the half-life of the profile.

[john-harrold]

Is it possible to extract these values from the output of PKNCA?

[billdenney]

Tlast and clast.pred are on the line. (Clast.obs is not necessarily on the line.)

That said, both of the shiny app developers are asking for this. So, @Gero1999 what are you doing now for plotting the half life, and do you both think that the intercept should be added as a parameter that comes out with half-life?

[Gero1999]

I actually was developing a function get_halflife_fit for the issue #339. I forgot that I stopped it because of #471 but now that is solved I can eventually retake it

In aNCA we have our own way of plotting using the PKNCAresults object. But the assumptions we use are only valid for the type of object we have. I am working on a generalized alternative that I will share with PKNCA soon :)

[john-harrold]

You know I missed the pred :). Sorry about that. I can use that much more easily :). Thankyou.

[Gero1999]

hey @john-harrold actually ruminate is really really cool! In aNCA we are doing something similar but exclusively for NCA. I don't know if there is any collaboration possible, but in case there is, would you be open in considering joining forces?

[john-harrold]

Hello Gerardo. I'm happy to work together. If there is something specific you have in mind please feel free to email me (john.m.harrold@gmail.com)