fixed allele names in simulation and io tests

.Rbuildignore

@@ -5,4 +5,5 @@ _libs/
 shell.nix
 .*sqlite
 .Rprofile
-conda.yml
+conda.yml
+zellij.kdl

R/io.R

@@ -552,8 +552,10 @@ fn_G_to_vcf = function(G, min_depth=100, max_depth=1000, verbose=FALSE) {
     }
     ### Extract the names of the alternative alleles
     if (is.null(G_alt)) {
-        vec_alleles_list = c("A", "T", "C", "G")
-        vec_alt = unlist(lapply(list_ids_chr_pos_all$vec_all, FUN=function(x){ sample(size=1, x=vec_alleles_list[grep(x, vec_alleles_list, ignore.case=TRUE, invert=TRUE)]) }))
+        # print("WARNING! Alternative alleles are unknown. Making psuedo-alleles.")
+        # vec_alleles_list = c("A", "T", "C", "G")
+        # vec_alt = unlist(lapply(list_ids_chr_pos_all$vec_all, FUN=function(x){ sample(size=1, x=vec_alleles_list[grep(x, vec_alleles_list, ignore.case=TRUE, invert=TRUE)]) }))
+        vec_alt = rep("N", times=ncol(G))
         # vec_alt = rep("allele_alt", times=ncol(G))
     } else {
         ### Convert the tabs in the loci-alleles names into dashes so as not to interfere with the VCF format
@@ -1512,7 +1514,7 @@ fn_filter_genotype = function(G, maf=0.01, sdev_min=0.0001,
                 )))
             return(error)
         }
-        if (sum(unique(list_ids_chr_pos_all$vec_all) %in% "allele_1") > 0) {
+        if (length(unique(list_ids_chr_pos_all$vec_all)) < 1) {
             error = methods::new("gpError",
                 code=252,
                 message=paste0(

tests/testthat/test-io.R

@@ -151,8 +151,8 @@ test_that("fn_filter_genotype", {
     expect_equal(sum(dim(G) == dim(G_filtered_1)), 2)
     ### Simulate SNP list for filtering
     G = fn_load_genotype(list_sim$fname_geno_vcf, retain_minus_one_alleles_per_locus=FALSE)
-    colnames(G) = gsub("allele_1", "A", colnames(G)) ### Rename allele_1 and allele_alt to A and T, respectively to allow filtering using a SNP list
-    colnames(G) = gsub("allele_alt", "T", colnames(G)) ### Rename allele_1 and allele_alt to A and T, respectively to allow filtering using a SNP list
+    colnames(G) = gsub("allele_1", "A", colnames(G)) ### Rename allele_1 and N to A and T, respectively to allow filtering using a SNP list
+    colnames(G) = gsub("N", "T", colnames(G)) ### Rename allele_1 and N to A and T, respectively to allow filtering using a SNP list
     n_sim_missing = 100
     mat_loci = matrix(unlist(strsplit(colnames(G), "\t")), byrow=TRUE, ncol=3)
     vec_loci = unique(paste0(mat_loci[,1], "\t", mat_loci[,2]))
@@ -174,8 +174,8 @@ test_that("fn_filter_genotype", {
     expect_equal(sum(abs(G_filtered_3 - G_filtered_2_split_G) < 1e-12), prod(dim(G_filtered_3)))
     ### Using additional filtering by sparsity (only works if G has missing data)
     G = fn_load_genotype(list_sim$fname_geno_vcf, min_depth=10, max_depth=500)
-    colnames(G) = gsub("allele_1", "A", colnames(G)) ### Rename allele_1 and allele_alt to A and T, respectively to allow filtering using a SNP list
-    colnames(G) = gsub("allele_alt", "T", colnames(G)) ### Rename allele_1 and allele_alt to A and T, respectively to allow filtering using a SNP list
+    colnames(G) = gsub("allele_1", "A", colnames(G)) ### Rename allele_1 and N to A and T, respectively to allow filtering using a SNP list
+    colnames(G) = gsub("N", "T", colnames(G)) ### Rename allele_1 and N to A and T, respectively to allow filtering using a SNP list
     G_filtered_4 = fn_filter_genotype(G=G, maf=0.05, max_n_alleles=1, fname_snp_list=fname_snp_list,
         max_sparsity_per_locus=0.5,
         frac_topmost_sparse_loci_to_remove=0.01,

zellij.kdl

@@ -0,0 +1,12 @@
+layout {
+	pane split_direction="vertical" {
+		pane split_direction="horizontal" {
+			pane
+			pane
+		}
+		pane
+	}
+	pane size=2 borderless=true {
+		plugin location="zellij:status-bar"
+	}
+}