README.md
In saviclab/savictools: Streamlined Data Manipulation for Pharmacometrics
savictools
Overview
savictools is a collection of functions designed to make the
pharmacometrics workflow faster, simpler, and more intuitive.
Data preparation
cohort() generates simulation datasets for NONMEM.expwt() computes expected weight for children under 5.sparse() detects sparse vs. intensive PK sampling occasions.tad() computes time after dose.zscores() computes z-scores for WHO’s anthropometric indicators.
Visualization
cr_plot() visualizes the clinical relevance of covariates.etacorr_cat() and etacorr_cont() plot ETA-covariate
correlations.pk_plot() creates PK concentration curves (“spaghetti plots”).VPC() runs the Perl-speaks-Nonmem (PsN) vpc command from R, and
is also a wrapper for xpose::VPC.
Workflow optimization
param_table() creates nicely formatted tables of parameter
estimates for single or multiple models.parse_all_sse() parses PsN sse-generated SSE_results.csv files
and combines them into dataframes summarizing parameter statistics
and error rates.
Installation
To install the development version from GitHub:
# install.packages("devtools")
devtools::install_github("saviclab/savictools")
Usage
library(savictools)
Data preparation
cohort() is used to generate NONMEM-ready datasets for clinical PK trials, either by sampling from real datasets or generating synthetic data.
set.seed(12345)
pop_example
#> # A tibble: 1,000 × 5
#> ID SEX AGE WT HT
#> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 2 38 13.9 95
#> 2 2 1 45 15.6 99
#> 3 3 2 40 12.2 87.6
#> 4 4 2 8 7.5 69
#> 5 5 2 50 11.6 91.5
#> 6 6 2 43 14.4 96.7
#> 7 7 2 1 3 48.4
#> 8 8 1 30 11 78.5
#> 9 9 1 36 12 86.2
#> 10 10 1 20 8.9 78
#> # … with 990 more rows
# 1. Sampling 20 individuals, above 10 kg and below 120 cm, with a fixed dose of
# 200 mg, observing every 4 hours for one day and dosing at times 0, 5, and 12.
# Note that the data has columns called "WT" and "HT".
inc <- "WT > 10 & HT < 120"
ot <- seq(0, 24, by = 4)
dt <- c(0, 5, 12)
cohort(
pop_example,
include = inc,
n = 20,
obs_times = ot,
dose_times = dt,
amt = 200
)
#> # A tibble: 200 × 9
#> ID TIME EVID AMT DV SEX AGE WT HT
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 19 0 1 200 0 1 47 15.1 92
#> 2 19 0 0 0 0 1 47 15.1 92
#> 3 19 4 0 0 0 1 47 15.1 92
#> 4 19 5 1 200 0 1 47 15.1 92
#> 5 19 8 0 0 0 1 47 15.1 92
#> 6 19 12 1 200 0 1 47 15.1 92
#> 7 19 12 0 0 0 1 47 15.1 92
#> 8 19 16 0 0 0 1 47 15.1 92
#> 9 19 20 0 0 0 1 47 15.1 92
#> 10 19 24 0 0 0 1 47 15.1 92
#> # … with 190 more rows
# 2. Simulating data. We assume WT and HT are normally distributed random
# variables, with means and standard deviations of 16 and 3.4 for WT and 132
# and 13.6 for HT.
p1 <-
list("WT" = list("rnorm", 16, 3.4),
"HT" = list("rnorm", 132, 13.6))
cohort(
param = p1,
include = inc,
n = 20,
pop_size = 1000,
obs_times = ot,
dose_times = dt,
amt = 200,
original_id = FALSE
)
#> # A tibble: 200 × 7
#> ID TIME EVID AMT DV WT HT
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 1 200 0 14.4 118.
#> 2 1 0 0 0 0 14.4 118.
#> 3 1 4 0 0 0 14.4 118.
#> 4 1 5 1 200 0 14.4 118.
#> 5 1 8 0 0 0 14.4 118.
#> 6 1 12 1 200 0 14.4 118.
#> 7 1 12 0 0 0 14.4 118.
#> 8 1 16 0 0 0 14.4 118.
#> 9 1 20 0 0 0 14.4 118.
#> 10 1 24 0 0 0 14.4 118.
#> # … with 190 more rows
# 3. As in (2), except we now define a dosing function.
dose_fun <- function(WT) {
ifelse(WT < 16, 150,
ifelse(WT < 20, 200, 250))
}
cohort(
param = p1,
include = inc,
n = 20,
pop_size = 500,
obs_times = ot,
dose_times = dt,
original_id = FALSE,
amt = dose_fun
)
#> # A tibble: 200 × 7
#> ID TIME EVID AMT DV WT HT
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 1 200 0 16.1 119.
#> 2 1 0 0 0 0 16.1 119.
#> 3 1 4 0 0 0 16.1 119.
#> 4 1 5 1 200 0 16.1 119.
#> 5 1 8 0 0 0 16.1 119.
#> 6 1 12 1 200 0 16.1 119.
#> 7 1 12 0 0 0 16.1 119.
#> 8 1 16 0 0 0 16.1 119.
#> 9 1 20 0 0 0 16.1 119.
#> 10 1 24 0 0 0 16.1 119.
#> # … with 190 more rows
set.seed(NULL)
expwt() calculates expected weight based on age and sex in under-5 children, adding a column called EXPWT.
expwt(pop_example)
#> # A tibble: 1,000 × 6
#> ID SEX AGE WT HT EXPWT
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 2 38 13.9 95 14.2
#> 2 2 1 45 15.6 99 15.8
#> 3 3 2 40 12.2 87.6 14.6
#> 4 4 2 8 7.5 69 7.9
#> 5 5 2 50 11.6 91.5 16.4
#> 6 6 2 43 14.4 96.7 15.2
#> 7 7 2 1 3 48.4 4.2
#> 8 8 1 30 11 78.5 13.3
#> 9 9 1 36 12 86.2 14.3
#> 10 10 1 20 8.9 78 11.3
#> # … with 990 more rows
sparse() automatically detects sparse vs. intensive PK sampling in a dataset.
library(ggplot2)
pk_example
#> # A tibble: 540 × 7
#> ID TIME AMT DV EVID ADDL II
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 50 0 4 0 0
#> 2 1 24.4 50 0 1 0 0
#> 3 1 48.3 50 0 1 0 0
#> 4 1 72.4 50 0 1 0 0
#> 5 1 96.0 50 0 1 0 0
#> 6 1 120. 50 0 1 0 0
#> 7 1 120. 0 0.815 0 0 0
#> 8 1 121. 0 9.48 0 0 0
#> 9 1 122. 0 10.4 0 0 0
#> 10 1 123. 0 10.1 0 0 0
#> # … with 530 more rows
sparse(pk_example, plot = TRUE)
sparse(pk_example, plot = TRUE) +
facet_wrap("ID", scales = "free_x", )
tad() computes time after dose.
# Deleting and re-calculating TAD
tad(pk_example)
#> # A tibble: 540 × 8
#> ID TIME AMT DV EVID ADDL II TAD
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 50 0 4 0 0 0
#> 2 1 24.4 50 0 1 0 0 0
#> 3 1 48.3 50 0 1 0 0 0
#> 4 1 72.4 50 0 1 0 0 0
#> 5 1 96.0 50 0 1 0 0 0
#> 6 1 120. 50 0 1 0 0 0
#> 7 1 120. 0 0.815 0 0 0 0.159
#> 8 1 121. 0 9.48 0 0 0 1.34
#> 9 1 122. 0 10.4 0 0 0 2.56
#> 10 1 123. 0 10.1 0 0 0 3.82
#> # … with 530 more rows
zscores() computes z-scores indicating nutritional status.
zscores(pop_example, units = "months")
#> # A tibble: 1,000 × 14
#> ID SEX AGE WT HT BMI HAZ WAZ WHZ BAZ HCZ ACZ TSZ
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <lgl> <lgl> <lgl>
#> 1 1 2 38 13.9 95 15.4 -0.37 -0.18 0.01 0.03 NA NA NA
#> 2 2 1 45 15.6 99 15.9 -0.64 -0.13 0.41 0.42 NA NA NA
#> 3 3 2 40 12.2 87.6 15.9 -2.55 -1.4 0.14 0.41 NA NA NA
#> 4 4 2 8 7.5 69 15.8 0.1 -0.48 -0.66 -0.75 NA NA NA
#> 5 5 2 50 11.6 91.5 13.9 -2.83 -2.63 -1.32 -1.06 NA NA NA
#> 6 6 2 43 14.4 96.7 15.4 -0.73 -0.38 0.06 0.07 NA NA NA
#> 7 7 2 1 3 48.4 12.8 -2.68 -2.33 -0.17 -1.3 NA NA NA
#> 8 8 1 30 11 78.5 17.9 -3.94 -1.62 0.78 1.52 NA NA NA
#> 9 9 1 36 12 86.2 16.1 -2.67 -1.48 0.03 0.44 NA NA NA
#> 10 10 1 20 8.9 78 14.6 -2.21 -2.16 -1.53 -1.15 NA NA NA
#> # … with 990 more rows, and 1 more variable: SSZ <lgl>
Visualization
saviclab/savictools documentation built on Dec. 7, 2023, 11:56 p.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
savictools
Overview
savictools is a collection of functions designed to make the pharmacometrics workflow faster, simpler, and more intuitive.
Data preparation
cohort()generates simulation datasets for NONMEM.expwt()computes expected weight for children under 5.sparse()detects sparse vs. intensive PK sampling occasions.tad()computes time after dose.zscores()computes z-scores for WHO’s anthropometric indicators.
Visualization
cr_plot()visualizes the clinical relevance of covariates.etacorr_cat()andetacorr_cont()plot ETA-covariate correlations.pk_plot()creates PK concentration curves (“spaghetti plots”).VPC()runs the Perl-speaks-Nonmem (PsN)vpccommand from R, and is also a wrapper for xpose::VPC.
Workflow optimization
param_table()creates nicely formatted tables of parameter estimates for single or multiple models.parse_all_sse()parses PsNsse-generated SSE_results.csv files and combines them into dataframes summarizing parameter statistics and error rates.
Installation
To install the development version from GitHub:
# install.packages("devtools")
devtools::install_github("saviclab/savictools")
Usage
library(savictools)
Data preparation
cohort() is used to generate NONMEM-ready datasets for clinical PK trials, either by sampling from real datasets or generating synthetic data.
set.seed(12345)
pop_example
#> # A tibble: 1,000 × 5
#> ID SEX AGE WT HT
#> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 2 38 13.9 95
#> 2 2 1 45 15.6 99
#> 3 3 2 40 12.2 87.6
#> 4 4 2 8 7.5 69
#> 5 5 2 50 11.6 91.5
#> 6 6 2 43 14.4 96.7
#> 7 7 2 1 3 48.4
#> 8 8 1 30 11 78.5
#> 9 9 1 36 12 86.2
#> 10 10 1 20 8.9 78
#> # … with 990 more rows
# 1. Sampling 20 individuals, above 10 kg and below 120 cm, with a fixed dose of
# 200 mg, observing every 4 hours for one day and dosing at times 0, 5, and 12.
# Note that the data has columns called "WT" and "HT".
inc <- "WT > 10 & HT < 120"
ot <- seq(0, 24, by = 4)
dt <- c(0, 5, 12)
cohort(
pop_example,
include = inc,
n = 20,
obs_times = ot,
dose_times = dt,
amt = 200
)
#> # A tibble: 200 × 9
#> ID TIME EVID AMT DV SEX AGE WT HT
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 19 0 1 200 0 1 47 15.1 92
#> 2 19 0 0 0 0 1 47 15.1 92
#> 3 19 4 0 0 0 1 47 15.1 92
#> 4 19 5 1 200 0 1 47 15.1 92
#> 5 19 8 0 0 0 1 47 15.1 92
#> 6 19 12 1 200 0 1 47 15.1 92
#> 7 19 12 0 0 0 1 47 15.1 92
#> 8 19 16 0 0 0 1 47 15.1 92
#> 9 19 20 0 0 0 1 47 15.1 92
#> 10 19 24 0 0 0 1 47 15.1 92
#> # … with 190 more rows
# 2. Simulating data. We assume WT and HT are normally distributed random
# variables, with means and standard deviations of 16 and 3.4 for WT and 132
# and 13.6 for HT.
p1 <-
list("WT" = list("rnorm", 16, 3.4),
"HT" = list("rnorm", 132, 13.6))
cohort(
param = p1,
include = inc,
n = 20,
pop_size = 1000,
obs_times = ot,
dose_times = dt,
amt = 200,
original_id = FALSE
)
#> # A tibble: 200 × 7
#> ID TIME EVID AMT DV WT HT
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 1 200 0 14.4 118.
#> 2 1 0 0 0 0 14.4 118.
#> 3 1 4 0 0 0 14.4 118.
#> 4 1 5 1 200 0 14.4 118.
#> 5 1 8 0 0 0 14.4 118.
#> 6 1 12 1 200 0 14.4 118.
#> 7 1 12 0 0 0 14.4 118.
#> 8 1 16 0 0 0 14.4 118.
#> 9 1 20 0 0 0 14.4 118.
#> 10 1 24 0 0 0 14.4 118.
#> # … with 190 more rows
# 3. As in (2), except we now define a dosing function.
dose_fun <- function(WT) {
ifelse(WT < 16, 150,
ifelse(WT < 20, 200, 250))
}
cohort(
param = p1,
include = inc,
n = 20,
pop_size = 500,
obs_times = ot,
dose_times = dt,
original_id = FALSE,
amt = dose_fun
)
#> # A tibble: 200 × 7
#> ID TIME EVID AMT DV WT HT
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 1 200 0 16.1 119.
#> 2 1 0 0 0 0 16.1 119.
#> 3 1 4 0 0 0 16.1 119.
#> 4 1 5 1 200 0 16.1 119.
#> 5 1 8 0 0 0 16.1 119.
#> 6 1 12 1 200 0 16.1 119.
#> 7 1 12 0 0 0 16.1 119.
#> 8 1 16 0 0 0 16.1 119.
#> 9 1 20 0 0 0 16.1 119.
#> 10 1 24 0 0 0 16.1 119.
#> # … with 190 more rows
set.seed(NULL)
expwt() calculates expected weight based on age and sex in under-5 children, adding a column called EXPWT.
expwt(pop_example)
#> # A tibble: 1,000 × 6
#> ID SEX AGE WT HT EXPWT
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 2 38 13.9 95 14.2
#> 2 2 1 45 15.6 99 15.8
#> 3 3 2 40 12.2 87.6 14.6
#> 4 4 2 8 7.5 69 7.9
#> 5 5 2 50 11.6 91.5 16.4
#> 6 6 2 43 14.4 96.7 15.2
#> 7 7 2 1 3 48.4 4.2
#> 8 8 1 30 11 78.5 13.3
#> 9 9 1 36 12 86.2 14.3
#> 10 10 1 20 8.9 78 11.3
#> # … with 990 more rows
sparse() automatically detects sparse vs. intensive PK sampling in a dataset.
library(ggplot2)
pk_example
#> # A tibble: 540 × 7
#> ID TIME AMT DV EVID ADDL II
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 50 0 4 0 0
#> 2 1 24.4 50 0 1 0 0
#> 3 1 48.3 50 0 1 0 0
#> 4 1 72.4 50 0 1 0 0
#> 5 1 96.0 50 0 1 0 0
#> 6 1 120. 50 0 1 0 0
#> 7 1 120. 0 0.815 0 0 0
#> 8 1 121. 0 9.48 0 0 0
#> 9 1 122. 0 10.4 0 0 0
#> 10 1 123. 0 10.1 0 0 0
#> # … with 530 more rows
sparse(pk_example, plot = TRUE)
sparse(pk_example, plot = TRUE) +
facet_wrap("ID", scales = "free_x", )
tad() computes time after dose.
# Deleting and re-calculating TAD
tad(pk_example)
#> # A tibble: 540 × 8
#> ID TIME AMT DV EVID ADDL II TAD
#> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 1 0 50 0 4 0 0 0
#> 2 1 24.4 50 0 1 0 0 0
#> 3 1 48.3 50 0 1 0 0 0
#> 4 1 72.4 50 0 1 0 0 0
#> 5 1 96.0 50 0 1 0 0 0
#> 6 1 120. 50 0 1 0 0 0
#> 7 1 120. 0 0.815 0 0 0 0.159
#> 8 1 121. 0 9.48 0 0 0 1.34
#> 9 1 122. 0 10.4 0 0 0 2.56
#> 10 1 123. 0 10.1 0 0 0 3.82
#> # … with 530 more rows
zscores() computes z-scores indicating nutritional status.
zscores(pop_example, units = "months")
#> # A tibble: 1,000 × 14
#> ID SEX AGE WT HT BMI HAZ WAZ WHZ BAZ HCZ ACZ TSZ
#> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <lgl> <lgl> <lgl>
#> 1 1 2 38 13.9 95 15.4 -0.37 -0.18 0.01 0.03 NA NA NA
#> 2 2 1 45 15.6 99 15.9 -0.64 -0.13 0.41 0.42 NA NA NA
#> 3 3 2 40 12.2 87.6 15.9 -2.55 -1.4 0.14 0.41 NA NA NA
#> 4 4 2 8 7.5 69 15.8 0.1 -0.48 -0.66 -0.75 NA NA NA
#> 5 5 2 50 11.6 91.5 13.9 -2.83 -2.63 -1.32 -1.06 NA NA NA
#> 6 6 2 43 14.4 96.7 15.4 -0.73 -0.38 0.06 0.07 NA NA NA
#> 7 7 2 1 3 48.4 12.8 -2.68 -2.33 -0.17 -1.3 NA NA NA
#> 8 8 1 30 11 78.5 17.9 -3.94 -1.62 0.78 1.52 NA NA NA
#> 9 9 1 36 12 86.2 16.1 -2.67 -1.48 0.03 0.44 NA NA NA
#> 10 10 1 20 8.9 78 14.6 -2.21 -2.16 -1.53 -1.15 NA NA NA
#> # … with 990 more rows, and 1 more variable: SSZ <lgl>
Visualization
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.