In fbzwsqualitasag/qzwsmix99rt:

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)

# library(qzwsmix99rt)
devtools::load_all()

Disclaimer

This package contains a collection of R-functions useful to run MiX99 from within R. The pre-requisite for this is that MiX99 is installed in a path that is reachable via the path environment variable.

First Example

This package contains an example with which the functionality can be tested. The following commands can be used to run the example analysis.

test_dir <- prepare_example_p1()

This copies the required files for the analysis into a temporary directory. The following shows a list of these files

list.files(path = test_dir)

The analysis is run with

s_log_file = file.path(test_dir, 
                       paste0(format(Sys.time(), "%Y%m%d%H%M%S"),
                              "_mix99.log", collapse = ""))
run_mix99(ps_work_dir = test_dir, ps_log_file = s_log_file)

The results and a logfile are written to the working directory

list.files(path = test_dir)

The content of the logfile can be inspected by

con_log <- file(description = s_log_file)
vec_log <- readLines(con = con_log)
close(con = con_log)
head(vec_log, 10)

The end of the logfile

tail(vec_log)

After the analysis all results can be removed by

fs::dir_delete(test_dir)

Conversion of QMSim Output to MiX99 Input

Given an output file from QMSim, the input files for MiX99 are generated. This is done by a conversion function. Currently this works only for simple animal models.

s_qm_path <- system.file("extdata", "qmsim", "p1_data_001.txt", package = "qzwsmix99rt")
(s_tmp_dir <- tempdir())
s_out_dir <- file.path(s_tmp_dir, 'mix99_work')
s_file_stem = paste0(format(Sys.time(), "%Y%m%d%H%M%S"),
                                                         "_mix99", collapse = "")
convert_qmsim_to_mix99(ps_qm_path = s_qm_path, 
                       ps_out_dir = s_out_dir, 
                       ps_file_stem = s_file_stem,
                       pl_vcmp      = list(genetic_variance = 0.1, residual_variance = 0.9))
list.files(path = s_out_dir)

The content of the generated data file can be inspected by

s_data_path <- file.path(s_out_dir, paste(s_file_stem, ".dat", sep = ""))
con_data <- file(s_data_path)
vec_data <- readLines(con = con_data)
close(con = con_data)
head(vec_data)
tail(vec_data)

The content of the pedigree can be viewed by

s_ped_path <- file.path(s_out_dir, paste(s_file_stem, ".ped", sep = ""))
con_ped <- file(s_ped_path)
vec_ped <- readLines(con = con_ped)
close(con = con_ped)
head(vec_ped)
tail(vec_ped)

The content of the parameter file

s_var_path <- file.path(s_out_dir, paste(s_file_stem, ".var", sep = ""))
con_var <- file(s_var_path)
vec_var <- readLines(con = con_var)
close(con = con_var)
cat(paste0(vec_var, collapse = "\n"), "\n")

The clim file

s_clm_path <- file.path(s_out_dir, paste(s_file_stem, ".clm", sep = ""))
con_clm <- file(s_clm_path)
vec_clm <- readLines(con = con_clm)
close(con = con_clm)
cat(paste0(vec_clm, collapse = "\n"), "\n")

Running MiX99 on Prepared Data

The data prepared so far are now used to run mix99

s_mix99_log_path <- file.path(s_out_dir,
                              paste0(format(Sys.time(), "%Y%m%d%H%M%S"),
                                     "_mix99.log", collapse = ""))
run_mix99(ps_work_dir = s_out_dir, ps_log_file = s_mix99_log_path)

The content of the working directory

list.files(path = s_out_dir)

The log file is

con_mix99_log <- file(s_mix99_log_path)
vec_mix99_log <- readLines(con = con_mix99_log)
close(con = con_mix99_log)
cat(paste0(vec_mix99_log, collapse = "\n"), "\n")

Problems with Recoding NA Values

tbl_qm <- read_qmsim(ps_path = s_qm_path)

(tbl_data <- dplyr::select(tbl_qm, Progeny, Sex, Phen))

(tbl_data <- dplyr::mutate(tbl_data, Sex = dplyr::recode(Sex, `M` = 1, `F` = 2)))

pn_mvc       = -99999.0
(tbl_data <- dplyr::mutate(tbl_data, Phen = tidyr::replace_na(Phen, pn_mvc)))

Clean Up

The generated files are removed by

fs::dir_delete(s_tmp_dir)

fbzwsqualitasag/qzwsmix99rt documentation built on April 2, 2022, 11:59 p.m.