Log in
Register|
yjlee168 ★★★   Kaohsiung, Taiwan, 2024-10-06 23:11 (106 d 19:19 ago) Posting: # 24218 Views: 1,329 |
|
|
Dear all, Due to retired/archived GTK2 packages (RGtk2, cairoDevice, gWidgets & gWidgetsRGtk2) since R v4.1.x, the installation of bear in all platforms was limited to R 4.1.x. The archived RGtk2 packages cannot run on ucrt64-based R (v4.2.x or above). Fortunately, since bear did not use cairoDevice package to output its graphics, users still can install it for R v4.2.x or above, including the current release (R v4.4,1). A while ago, Helmut suggested an installation method for bear running with R v4.2.x or above. To simplify the installation procedure, I wrote an R script (bw_preinst.r) to facilitate the process. Please note that the script currently was only applied for Windows 10/11 (I tested the script using R v4.4.1 under Virtualbox).
That's it. Good luck. Ps1. These four retired GTK2 packages (RGtk2, cairoDevice, gWidgets & gWidgetsRGtk2) were all compiled from the previous release (R 4.1.3) under Windows 10; Since they were all retired/archived, R will not upgrade these packages; They still work well. Ps2. I could not put any web link for this post; otherwise, I got Error 406  .— All the best, -- Yung-jin Lee bear v2.9.2:- created by Hsin-ya Lee & Yung-jin Lee Kaohsiung, Taiwan https://www.pkpd168.com/bear Download link (updated) -> here |
|
roman_max ☆ Russia, 2024-11-02 01:05 (80 d 16:24 ago) @ yjlee168 Posting: # 24254 Views: 864 |
|
|
Dear Yung-jin Lee, thank you for regular updates of bear package, it works fine on my PC whithout any problem (R 4.4.1, bear 2.9.1). Recently looking through the PK conc-time profiles I found that for some cases there were non-zero concentration on the plots, but in dataframe all pre-dose samples were 0. If you run demo data you also can see this. Is it possible to "force" the plot start from 0,0 exactly to avoid impression that non-zero (above LLOQ) values exist in pre-dose? |
|
yjlee168 ★★★ Kaohsiung, Taiwan, 2024-11-13 20:07 (68 d 21:22 ago) @ roman_max Posting: # 24273 Views: 665 |
|
|
Hi Roman, As we have discussed via email. Here is the final as follows. Just leave a note at the Forum. To set the origin of the plot to (0.,0.), some plots can be a little bit ugly or weird (esp. semi-log plots). First, the semi-log plots cannot be set the origin as (0.,0.). All we can do is set the origin as (0.,0.001) for example. Meanwhile, if we need to set the origin as (0.,0.), we also need to set the maximum values for the x- & y-axis simultaneously. It was easy to set the x-axis; however, it can be difficult for the y-axis, especially when plotting [mean ± sd] (regarding the size of error bars for each point; and it can be different from project to project; even from point to point in the same project) for drug plasma conc. no matter what it was a linear or semi-log plot. For now, I will leave bear as it was (aka. let ggplot() from ggplot2 package to decide). However, I will keep thinking about this issue.— All the best, -- Yung-jin Lee bear v2.9.2:- created by Hsin-ya Lee & Yung-jin Lee Kaohsiung, Taiwan https://www.pkpd168.com/bear Download link (updated) -> here |
Ing. Helmut Schütz