bguiastr
commented
2 years ago Hello,
Yes xpose's strength is that it consolidates the NONMEM outputs into a tidy format i.e. the xpdb. From the xpdb you can access the content of the NONMEM tables, files and code.
I have created a small example to show you how you can access the files of interest for you:
library(xpose)
#> Loading required package: ggplot2
#>
#> Attaching package: 'xpose'
#> The following object is masked from 'package:stats':
#>
#> filter
## Check which files are available in the xpdb
list_files(xpdb_ex_pk)
#> Files:
#> name extension problem subprob method data modified
#> <chr> <chr> <dbl> <dbl> <chr> <list> <lgl>
#> 1 run001.cor cor 1 0 foce <tibble [14 × 15]> FALSE
#> 2 run001.cov cov 1 0 foce <tibble [14 × 15]> FALSE
#> 3 run001.ext ext 1 0 foce <tibble [28 × 16]> FALSE
#> 4 run001.grd grd 1 0 foce <tibble [21 × 11]> FALSE
#> 5 run001.phi phi 1 0 foce <tibble [74 × 12]> FALSE
#> 6 run001.shk shk 1 0 foce <tibble [7 × 5]> FALSE
## Get the correlation matrix
get_file(xpdb_ex_pk, ext = "cor")
#> Returning data from run001.cor, $prob no.1, subprob no.0, method foce
#> # A tibble: 14 × 15
#> NAME THETA1 THETA2 THETA3 THETA4 THETA5 THETA6 THETA7 SIGMA(1…¹
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 THETA1 0.891 0.525 0.100 -0.246 0.0435 0.193 -0.218 0
#> 2 THETA2 0.525 0.0438 -0.234 -0.0932 0.257 -0.0198 0.0734 0
#> 3 THETA3 0.100 -0.234 0.809 0.253 -0.259 -0.220 -0.229 0
#> 4 THETA4 -0.246 -0.0932 0.253 0.0157 -0.164 -0.266 0.0924 0
#> 5 THETA5 0.0435 0.257 -0.259 -0.164 0.0224 -0.383 0.0827 0
#> 6 THETA6 0.193 -0.0198 -0.220 -0.266 -0.383 0.00366 -0.255 0
#> 7 THETA7 -0.218 0.0734 -0.229 0.0924 0.0827 -0.255 0.00170 0
#> 8 SIGMA(1,1) 0 0 0 0 0 0 0 0
#> 9 OMEGA(1,1) -0.115 -0.145 0.165 -0.139 -0.293 0.246 -0.0780 0
#> 10 OMEGA(2,1) 0 0 0 0 0 0 0 0
#> 11 OMEGA(2,2) -0.111 -0.0232 -0.131 -0.0750 -0.366 0.222 0.0787 0
#> 12 OMEGA(3,1) 0 0 0 0 0 0 0 0
#> 13 OMEGA(3,2) 0 0 0 0 0 0 0 0
#> 14 OMEGA(3,3) 0.155 -0.296 0.829 0.204 -0.228 0.00578 -0.268 0
#> # … with 6 more variables: `OMEGA(1,1)` <dbl>, `OMEGA(2,1)` <dbl>,
#> # `OMEGA(2,2)` <dbl>, `OMEGA(3,1)` <dbl>, `OMEGA(3,2)` <dbl>,
#> # `OMEGA(3,3)` <dbl>, and abbreviated variable name ¹`SIGMA(1,1)`
## Get the covariance matrix
get_file(xpdb_ex_pk, ext = "cov")
#> Returning data from run001.cov, $prob no.1, subprob no.0, method foce
#> # A tibble: 14 × 15
#> NAME THETA1 THETA2 THETA3 THETA4 THETA5 THETA6 THETA7 SIGMA…¹
#> <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 THETA1 7.95e-1 2.05e-2 7.22e-2 -3.45e-3 8.71e-4 6.30e-4 -3.30e-4 0
#> 2 THETA2 2.05e-2 1.92e-3 -8.30e-3 -6.42e-5 2.53e-4 -3.17e-6 5.46e-6 0
#> 3 THETA3 7.22e-2 -8.30e-3 6.55e-1 3.22e-3 -4.71e-3 -6.52e-4 -3.15e-4 0
#> 4 THETA4 -3.45e-3 -6.42e-5 3.22e-3 2.47e-4 -5.79e-5 -1.53e-5 2.46e-6 0
#> 5 THETA5 8.71e-4 2.53e-4 -4.71e-3 -5.79e-5 5.04e-4 -3.14e-5 3.15e-6 0
#> 6 THETA6 6.30e-4 -3.17e-6 -6.52e-4 -1.53e-5 -3.14e-5 1.34e-5 -1.58e-6 0
#> 7 THETA7 -3.30e-4 5.46e-6 -3.15e-4 2.46e-6 3.15e-6 -1.58e-6 2.88e-6 0
#> 8 SIGMA… 0 0 0 0 0 0 0 0
#> 9 OMEGA… -1.29e-3 -7.97e-5 1.68e-3 -2.75e-5 -8.26e-5 1.13e-5 -1.66e-6 0
#> 10 OMEGA… 0 0 0 0 0 0 0 0
#> 11 OMEGA… -1.23e-3 -1.27e-5 -1.33e-3 -1.47e-5 -1.03e-4 1.02e-5 1.67e-6 0
#> 12 OMEGA… 0 0 0 0 0 0 0 0
#> 13 OMEGA… 0 0 0 0 0 0 0 0
#> 14 OMEGA… 7.69e-2 -7.23e-3 3.74e-1 1.79e-3 -2.85e-3 1.18e-5 -2.54e-4 0
#> # … with 6 more variables: `OMEGA(1,1)` <dbl>, `OMEGA(2,1)` <dbl>,
#> # `OMEGA(2,2)` <dbl>, `OMEGA(3,1)` <dbl>, `OMEGA(3,2)` <dbl>,
#> # `OMEGA(3,3)` <dbl>, and abbreviated variable name ¹`SIGMA(1,1)`Created on 2022-10-21 with reprex v2.0.2
Dear XPOSE developer,
get_prm() and prm_table read "OMEGA - CORR MATRIX FOR RANDOM EFFECTS - ETAS" rather than "OMEGA - COV MATRIX FOR RANDOM EFFECTS - ETAS" by default.
Is there a way to read OMEGA - COV MATRIX FOR RANDOM EFFECTS - ETAS using XPOSE functions?
Thanks!