call_gt {bbmix}

Call genotypes using beta binomial after model training

Description

Call genotypes using beta binomial after model training

Usage

call_gt(
  allele_counts_f,
  depth = 10,
  stan_f = NULL,
  legend_f,
  pop = "EUR",
  prob = 0.99,
  fisher_f = NULL,
  fisher = 30,
  cluster_f = NULL,
  out
)

Arguments

allele_counts_f

vector with file names with allele counts for SNPs

depth

min read count to call variant

stan_f

full name to stan object with model fit to extract mean of parameters. Defaults to the model trained with genome in a bottle reads. Otherwise this object can be generated with fit_bb function.

legend_f

full name for file with SNP info to get allele frequency for prior

pop

population to select AF for GT prior, defaults to EUR

prob

cut-off for making hard calls, defaults to 0.99

fisher_f

file with Fisher test to detect strand bias

fisher

cut_off for Fisher test to detect strand bias

cluster_f

file with info about SNP clusters

out

character with file name to save genotype output

Value

data table with genotype probabilities

Examples

## Retrive input files for running call_gt
counts_f <- system.file("extdata/input", "NA12878.chr22.Q20.allelicCounts.txt",
package = "bbmix",
mustWork = TRUE)

legend <- system.file("extdata/input", "1000GP_Phase3_chr22.legend",
package = "bbmix", mustWork = TRUE)

fisher_f <- system.file("extdata/input", "chr22.FS.Q20.alleleCounts.txt",
package = "bbmix", mustWork = TRUE)

cluster_f <- system.file("extdata/input", "fSNPs_22_RP_maf0_01_cluster3window35.txt",
package = "bbmix", mustWork = TRUE)

out <- paste0(tempdir() , "/NA12878.chrom22.gt.txt")

## Run call_gt:
call_gt(allele_counts_f = counts_f,
legend_f = legend,
fisher_f = fisher_f,
cluster_f = cluster_f,
out = out)

unlink(out)