plinkQC (version 0.2.2)

evaluate_check_het_and_miss: Evaluate results from PLINK missing genotype and heterozygosity rate check.

Description

Evaluates and depicts results from plink --missing (missing genotype rates per individual) and plink --het (heterozygosity rates per individual) via run_check_heterozygosity and run_check_missingness or externally conducted check.) Non-systematic failures in genotyping and outlying heterozygosity (hz) rates per individual are often proxies for DNA sample quality. Larger than expected heterozygosity can indicate possible DNA contamination. The mean heterozygosity in PLINK is computed as hz_mean = (N-O)/N, where N: number of non-missing genotypes and O:observed number of homozygous genotypes for a given individual. Mean heterozygosity can differ between populations and SNP genotyping panels. Within a population and genotyping panel, a reduced heterozygosity rate can indicate inbreeding - these individuals will then be returned by check_relatedness as individuals that fail the relatedness filters. evaluate_check_het_and_miss creates a scatter plot with the individuals' missingness rates on x-axis and their heterozygosity rates on the y-axis.

Usage

evaluate_check_het_and_miss(qcdir, name, imissTh = 0.03, hetTh = 3,
  label = TRUE, interactive = FALSE)

Arguments

qcdir

[character] path/to/directory/with/QC/results containing name.imiss and name.het results as returned by plink --missing and plink --het.

name

[character] Prefix of PLINK files, i.e. name.bed, name.bim, name.fam, name.het and name.imiss.

imissTh

[double] Threshold for acceptable missing genotype rate in any individual; has to be proportion between (0,1)

hetTh

[double] Threshold for acceptable deviation from mean heterozygosity in any individual. Expressed as multiples of standard deviation of heterozygosity (het), i.e. individuals outside mean(het) +/- hetTh*sd(het) will be returned as failing heterozygosity check; has to be larger than 0.

label

[logical] Set TRUE, to add fail IDs as text labels in scatter plot.

interactive

[logical] Should plots be shown interactively? When choosing this option, make sure you have X-forwarding/graphical interface available for interactive plotting. Alternatively, set interactive=FALSE and save the returned plot object (p_het_imiss) via ggplot2::ggsave(p=p_het_imiss , other_arguments) or pdf(outfile) print(p_het_imiss) dev.off().

Value

named [list] with i) fail_imiss dataframe containing FID (Family ID), IID (Within-family ID), MISS_PHENO (Phenotype missing? (Y/N)), N_MISS (Number of missing genotype call(s), not including obligatory missings), N_GENO ( Number of potentially valid call(s)), F_MISS (Missing call rate) of individuals failing missing genotype check and ii) fail_het dataframe containing FID (Family ID), IID (Within-family ID), O(HOM) (Observed number of homozygotes), E(HOM) (Expected number of homozygotes), N(NM) (Number of non-missing autosomal genotypes), F (Method-of-moments F coefficient estimate) of individuals failing outlying heterozygosity check and iii) p_het_imiss, a ggplot2-object 'containing' a scatter plot with the samples' missingness rates on x-axis and their heterozygosity rates on the y-axis, which can be shown by print(p_het_imiss).

Details

All, run_check_heterozygosity, run_check_missingness and evaluate_check_het_and_miss can simply be invoked by check_het_and_miss.

For details on the output data.frame fail_imiss and fail_het, check the original description on the PLINK output format page: https://www.cog-genomics.org/plink/1.9/formats#imiss and https://www.cog-genomics.org/plink/1.9/formats#het

Examples

# NOT RUN {
qcdir <- system.file("extdata", package="plinkQC")
name <- "data"
# }
# NOT RUN {
fail_het_miss <- evaluate_check_het_and_miss(qcdir=qcdir, name=name,
interactive=FALSE)
# }