bug in seqMerge()

zhengxwen · Jun 22, 2024 · 702488b · 702488b
1 parent 661eb0a
commit 702488b
Show file tree

Hide file tree

Showing 3 changed files with 15 additions and 6 deletions.
diff --git a/DESCRIPTION b/DESCRIPTION
@@ -1,8 +1,8 @@
 Package: SeqArray
 Type: Package
 Title: Data Management of Large-Scale Whole-Genome Sequence Variant Calls
-Version: 1.44.0
-Date: 2024-04-27
+Version: 1.44.1
+Date: 2024-06-22
 Depends: R (>= 3.5.0), gdsfmt (>= 1.31.1)
 Imports: methods, parallel, IRanges, GenomicRanges, GenomeInfoDb, Biostrings,
         S4Vectors

diff --git a/NEWS b/NEWS
@@ -1,3 +1,12 @@
+CHANGES IN VERSION 1.44.1
+-------------------------
+
+BUG FIXES
+
+    o `seqMerge()` should internally use "chr_position_ref_alt" to distinguish
+      the variants in different files
+
+
 CHANGES IN VERSION 1.44.0
 -------------------------
 

diff --git a/R/UtilsMerge.R b/R/UtilsMerge.R
@@ -36,7 +36,7 @@
     .append_gds(n, flist, varnm, verbose)
     .DigestCode(n, digest, verbose)
     if (nVariant != objdesp.gdsn(n)$dim)
-        stop(sprintf("Invalid number of variants in '%s'.", varnm))
+        warning(sprintf("Invalid number of variants in '%s'.", varnm), immediate.=TRUE)
     n
 }
 
@@ -319,15 +319,15 @@ seqMerge <- function(gds.fn, out.fn, storage.option="LZMA_RA",
             nVariant <- nVariant + objdesp.gdsn(index.gdsn(f, "variant.id"))$dim
         }
     } else {
-        variant.id <- variant2.id <- seqGetData(flist[[1L]], "$chrom_pos2")
+        variant.id <- variant2.id <- seqGetData(flist[[1L]], "$chrom_pos_allele")
         if (verbose)
         {
             cat(sprintf("    [%-2d] %s (%s variant%s)\n", 1L, basename(gds.fn[1L]),
                 .pretty(length(variant.id)), .plural(length(variant.id))))
         }
         for (i in seq_along(flist)[-1L])
         {
-            s <- seqGetData(flist[[i]], "$chrom_pos2")
+            s <- seqGetData(flist[[i]], "$chrom_pos_allele")
             if (verbose)
             {
                 cat(sprintf("    [%-2d] %s (%s variant%s)\n", i,
@@ -366,7 +366,7 @@ seqMerge <- function(gds.fn, out.fn, storage.option="LZMA_RA",
         varidx <- vector("list", length(flist))
         for (i in seq_along(flist))
         {
-            varidx[[i]] <- match(seqGetData(flist[[i]], "$chrom_pos2"),
+            varidx[[i]] <- match(seqGetData(flist[[i]], "$chrom_pos_allele"),
                 variant.id)
             if (is.unsorted(varidx[[i]], strictly=TRUE))
                 stop("File ", i, ": chromosomes and positions are unsorted.")