Vcf windows

R/gen_tibble.R

@@ -3,7 +3,17 @@
 #' A `gen_tibble` stores genotypes for individuals in a tidy format. DESCRIBE
 #' here the format
 #'
-#' When loading packedancestry files, missing alleles will be converted from
+#'- *VCF* files: the fast `cpp` parser is used by default. Both `cpp` and `vcfR` parsers
+#' attempt to establish ploidy from the first variant; if that variant is found in a
+#' sex chromosome (or mtDNA), the parser will fail with 'Error: a genotype has more
+#' than max_ploidy alleles...'. To use the fast parser, change the order of variants
+#' so that the first chromosome is an autosome using a tool such as `vcftools`.
+#' Currently, only biallelic SNPs are supported. If haploid variants (e.g. sex
+#' chromosomes) are included in the vcf, they are not transformed into homozygous
+#' calls. Instead, reference alleles will be counted as 0 and alternative alleles
+#' will be counted as 1.
+#'
+#' - *packedancestry* files: When loading *packedancestry* files, missing alleles will be converted from
 #' 'X' to NA
 #'
 #' @param x can be:

R/vcf_to_fbm_vcfR.R

@@ -157,7 +157,7 @@ poly_genotype_dosage <- function (x, max_ploidy){
     if (dosage<(max_ploidy+1)){
       return(as.raw(dosage))
     } else{
-      stop("a genotype has more than max_ploidy alleles. We estimate max_plody from the first variant in the vcf file, make sure that variant is representative of ploidy (e.g. it is not on a sex chromosome).")
+      stop("a genotype has more than max_ploidy alleles. We estimate max_ploidy from the first variant in the vcf file, make sure that variant is representative of ploidy (e.g. it is not on a sex chromosome).")
     }
   } else {
     return(as.raw(max_ploidy+1))

src/vcf_parser.cpp

@@ -43,7 +43,7 @@ int countAlternateAlleles(const std::string &genotype, const size_t missingValue
     //   Rcout<<pos_i<<"\t"<<missingValue<<std::endl;
     // }
     if (pos_i> ((missingValue-1)*2)){
-      Rcpp::stop("a genotype has more than max_ploidy alleles. We estimate max_plody from the first variant in the vcf file, make sure that variant is representative of ploidy (e.g. it is not on a sex chromosome).");
+      Rcpp::stop("a genotype has more than max_ploidy alleles. We estimate max_ploidy from the first variant in the vcf file, make sure that variant is representative of ploidy (e.g. it is not on a sex chromosome).");
     }
 
     //

tests/testthat/test_gen_tibble.R

@@ -638,7 +638,7 @@ test_that("vcf's with haploid markers first give errors",{
   # the cpp parser catches the problem
   expect_error(pop_a_vcf_gt_hap_cpp <- gen_tibble(vcf_path_haploid, quiet=TRUE,backingfile = tempfile(), parser="cpp"),
                "a genotype")
-  # vcfR fails to raise an error
+  # vcfR catches the problem
   expect_error(pop_a_vcf_gt_hap_vcfR <- gen_tibble(vcf_path_haploid, quiet=TRUE,backingfile = tempfile(), parser="vcfR"),
                   "a genotype")
 })