Description Usage Arguments Details Value Examples
Run snptest gwas on multiple cores
1 2 3 4 |
indir |
Directory with gz-compressed chunk files. |
sample_file |
Phenotype file. |
outdir |
Directory where snptest output files should go. |
pheno |
Name of phenotype variable as used in ‘sample_file’. |
covs |
Character vector with names of covariates as used in ‘sample_file’. |
exclusion_file |
File with ids of individuals to be excluded from analysis. |
add_args |
Additional command-line arguments to snptest. |
ncore |
Number of cores to use in parallel. |
pattern |
Regex pattern used to match input chunk files in ‘indir’. |
chr_chunk |
Extended regular expression with two parenthesized subexpressions matching chromosome and chunk number in input chunk file names. |
executable |
Path to snptest executable. |
Function parameters without default values are mandatory. All other parameters are optional.
Sometimes you do not want all chunk files in ‘indir’ to be analyzed.
A typical reason would be that different chunk files need to be run
with different snptest options. If you are able to specify a regular
expression that matches the set of chunk files that you do want to
analyze, you can supply this regular expression as the ‘pattern’
argument to snptest
. This will cause snptest to be run only
with files that are located in ‘indir’ _and_ whose path—which
includes ‘indir’ as a prefix (!)—matches the regular expression
‘pattern’. Make sure that your pattern a) matches only chunk files
and b) that all matched chunk files can be snptest'ed using the same
set of snptest options.
If your chunk files follow a naming scheme where chromosome and chunk number cannot be matched by the default regular expression for ‘chr_chunk’, then you will have to specify your own regular expression (see regex). This (extended) regular expression must contain two parenthesized subexpressions. The first must match the chromosome, the second the chunk part of your chunk file names.
Returns a data frame with columns: ‘chr’ (chromosome number), ‘chunk’ (chunk number), ‘input’ (gz-compressed input file), ‘output’ (snptest output file), ‘log’ (snptest log file), ‘done’ (TRUE for successfully snptest'ed chunks, otherwise FALSE).
After an entirely successful run of snptest
all values
in the ‘done’ column will be "TRUE". Otherwise columns with 'done ==
"FALSE"' should give you a hint where to look for possible problems.
Also take a look at the warning, if any, that snptest
emits while running.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 | library(genFun)
## Not run:
## Run snptest on men-only nonPAR chunks on chromosome X.
##
## Note: - The sample file must contain the same individuals as the
## imputed chunk files.
## - You have to specify "-assume_chromosome X".
## - You have to craft your regular expression for `chr_chunk'
## because the nonPAR chunk files don't follow the default
## naming convention where chromosome and chunk are specified as
## in "chr3_7".
## - You have to specify a custom `pattern' because you need to
## restrict the set of chunk files to those that contain only
## male subjects.
snptest(
indir = "path/to/directory/with/gz-compressed/input/files",
sample_file = "path/to/MEN_ONLY_phenotype_file",
outdir = "path/to/directory/for/snptest/output/files",
pheno = "blood_sugar",
covs = c("age", "bmi"),
exclusion_file = "path/to/file/with/exclusion/ids",
add_args = c("-missing_code NA",
"-frequentist 1",
"-method expected",
"-hwe",
"-printids",
"-use_raw_covariates",
"-use_raw_phenotypes",
"-assume_chromosome XY"),
ncore = 9L,
pattern = "males.*\\.gz$",
chr_chunk = ".*chr([^_]+)_(nonPAR_\\d+)",
executable = "/opt/snptest_v2.5_linux_x86_64_static/snptest_v2.5")
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.