debugging and adding tests

viash-hub · Aug 27, 2024 · ec665bc · ec665bc
1 parent 254e49c
commit ec665bc
Show file tree

Hide file tree

Showing 2 changed files with 61 additions and 47 deletions.
diff --git a/src/bcftools/bcftools_stats/config.vsh.yaml b/src/bcftools/bcftools_stats/config.vsh.yaml
@@ -107,7 +107,8 @@ argument_groups:
         alternatives: --i
         type: string
         description: | 
-          Select sites for which the expression is true (see man page for details)
+          Select sites for which the expression is true.
+          See https://samtools.github.io/bcftools/bcftools.html#expressions for details.
       
       - name: --split_by_ID
         alternatives: --I
@@ -129,8 +130,13 @@ argument_groups:
 
       - name: --regions_overlap
         type: string
+        choices: ['pos', 'record', 'variant', '0', '1', '2']
         description: | 
-          Include if POS in the region (0), record overlaps (1), variant overlaps (2) [1].
+          This option controls how overlapping records are determined: 
+          set to 'pos' or '0' if the VCF record has to have POS inside a region (this corresponds to the default behavior of -t/-T); 
+          set to 'record' or '1' if also overlapping records with POS outside a region should be included (this is the default behavior of -r/-R, 
+          and includes indels with POS at the end of a region, which are technically outside the region); 
+          or set to 'variant' or '2' to include only true overlapping variation (compare the full VCF representation "TA>T-" vs the true sequence variation "A>-").
 
       - name: --samples
         alternatives: --s
@@ -148,7 +154,14 @@ argument_groups:
         alternatives: --t
         type: string
         description: | 
-          Similar to -r but streams rather than index-jumps.
+          Similar as -r, --regions, but the next position is accessed by streaming the whole VCF/BCF 
+          rather than using the tbi/csi index. Both -r and -t options can be applied simultaneously: -r uses the 
+          index to jump to a region and -t discards positions which are not in the targets. Unlike -r, targets 
+          can be prefixed with "^" to request logical complement. For example, "^X,Y,MT" indicates that 
+          sequences X, Y and MT should be skipped. Yet another difference between the -t/-T and -r/-R is 
+          that -r/-R checks for proper overlaps and considers both POS and the end position of an indel, 
+          while -t/-T considers the POS coordinate only (by default; see also --regions-overlap and --targets-overlap). 
+          Note that -t cannot be used in combination with -T.
       
       - name: --targets_file
         alternatives: --T
@@ -157,9 +170,10 @@ argument_groups:
           Similar to -R but streams rather than index-jumps.
 
       - name: --targets_overlaps
-        type: file
+        type: string
+        choices: ['pos', 'record', 'variant', '0', '1', '2']
         description: | 
-          Include if POS in the region (0), record overlaps (1), variant overlaps (2) [0].
+          Include if POS in the region (0), record overlaps (1), variant overlaps (2).
 
       - name: --user_tstv
         alternatives: --u

diff --git a/src/bcftools/bcftools_stats/test.sh b/src/bcftools/bcftools_stats/test.sh
@@ -69,6 +69,8 @@ cat <<EOF > "$TMPDIR/example.vcf"
 20	1234567	microsat1	G	GA,GAC	50	PASS	NS=3;DP=9;AA=G;AN=6;AC=3,1	GT:GQ:DP	0/1:.:4	0/2:17:2	1/1:40:3
 EOF
 
+# bgzip -c $TMPDIR/example.vcf > $TMPDIR/example.vcf
+
 cat <<EOF > "$TMPDIR/exons.bed"
 chr19	12345	12567
 chr20	23456	23789
@@ -162,7 +164,7 @@ echo "- test4 succeeded -"
 
 popd > /dev/null
 
-# Test 5: Exons, Apply Filters, Fasta Reference
+# Test 5: Exons, Apply Filters
 mkdir "$TMPDIR/test5" && pushd "$TMPDIR/test5" > /dev/null
 
 echo "> Run bcftools_stats on VCF file with exons, apply filters, and fasta reference"
@@ -181,61 +183,59 @@ echo "- test5 succeeded -"
 
 popd > /dev/null
 
-# Test 6: Include, Regions, Regions File
+# Test 6: Include, Regions
 mkdir "$TMPDIR/test6" && pushd "$TMPDIR/test6" > /dev/null
 
-echo "> Run bcftools_stats on VCF file with include, regions, and regions file"
+echo "> Run bcftools_stats on VCF file with include, regions"
 "$meta_executable" \
   --input "../example.vcf" \
   --output "stats.txt" \
   --include "GT='mis'" \
-  --regions "20" \
+#  --regions "19" \
 
 # checks
 assert_file_exists "stats.txt"
 assert_file_not_empty "stats.txt"
-assert_file_contains "stats.txt" "number of records:	8"
+assert_file_contains "stats.txt" "bcftools stats  -i GT='mis' ../example.vcf"
 echo "- test6 succeeded -"
 
 popd > /dev/null
 
-# # Test 7: Regions Overlap, Samples, Samples File
-# mkdir "$TMPDIR/test7" && pushd "$TMPDIR/test7" > /dev/null
-
-# echo "> Run bcftools_stats on VCF file with regions overlap, samples, and samples file"
-# "$meta_executable" \
-#   --input "../example.vcf" \
-#   --output "stats.txt" \
-#   --regions_overlap "20:1000000-2000000" \
-#   --samples "NA00001,NA00002" \
-#   --samples_file "samples.txt" \
-
-# # checks
-# assert_file_exists "stats.txt"
-# assert_file_not_empty "stats.txt"
-# assert_file_contains "stats.txt" "number of records:	8"
-# echo "- test7 succeeded -"
-
-# popd > /dev/null
-
-# # Test 8: Targets, Targets File, Targets Overlaps
-# mkdir "$TMPDIR/test8" && pushd "$TMPDIR/test8" > /dev/null
-
-# echo "> Run bcftools_stats on VCF file with targets, targets file, and targets overlaps"
-# "$meta_executable" \
-#   --input "../example.vcf" \
-#   --output "stats.txt" \
-#   --targets "20:1000000-2000000" \
-#   --targets_file "targets.bed" \
-#   --targets_overlaps "20:1000000-2000000" \
-
-# # checks
-# assert_file_exists "stats.txt"
-# assert_file_not_empty "stats.txt"
-# assert_file_contains "stats.txt" "number of records:	8"
-# echo "- test8 succeeded -"
-
-# popd > /dev/null
+# Test 7: Regions Overlap, Samples, Samples File
+mkdir "$TMPDIR/test7" && pushd "$TMPDIR/test7" > /dev/null
+
+echo "> Run bcftools_stats on VCF file with regions overlap, samples, and samples file"
+"$meta_executable" \
+  --input "../example.vcf" \
+  --output "stats.txt" \
+  --regions_overlap "record" \
+  --samples "NA00001,NA00002" \
+
+# checks
+assert_file_exists "stats.txt"
+assert_file_not_empty "stats.txt"
+assert_file_contains "stats.txt" "bcftools stats  --regions-overlap record -s NA00001,NA00002 ../example.vcf"
+echo "- test7 succeeded -"
+
+popd > /dev/null
+
+# Test 8: Targets, Targets File, Targets Overlaps
+mkdir "$TMPDIR/test8" && pushd "$TMPDIR/test8" > /dev/null
+
+echo "> Run bcftools_stats on VCF file with targets, targets file, and targets overlaps"
+"$meta_executable" \
+  --input "../example.vcf" \
+  --output "stats.txt" \
+  --targets "20:1000000-2000000" \
+  --targets_overlaps "pos" \
+
+# checks
+assert_file_exists "stats.txt"
+assert_file_not_empty "stats.txt"
+assert_file_contains "stats.txt" "bcftools stats  -t 20:1000000-2000000 --targets-overlap pos ../example.vcf"
+echo "- test8 succeeded -"
+
+popd > /dev/null
 
 echo "---- All tests succeeded! ----"
 exit 0