Merge branch 'dev' of git.illumina.com:Bioinformatics/hap.py into dev

RELEASES.md

@@ -1,5 +1,14 @@
 # Hap.py Release Notes / Change Log
 
+## v0.3.3
+
+| Ticket  | Description                                                                        |
+|---------|------------------------------------------------------------------------------------|
+| HAP-232 | Fix help text in hap.py and pre.py                                                 |
+| HAP-233 | INFO fields don't get mixed up in preprocessing anymore                            |
+| HAP-234 | Forward unstructured headers through hap.py                                        |
+| HAP-236 | Fix quanitify test to work with compressed outputs                                 |
+
 ## v0.3.2
 
 | Ticket  | Description                                                                        |

src/c++/lib/variant/VariantAlleleSplitter.cpp

@@ -315,6 +315,7 @@ bool VariantAlleleSplitter::advance()
                 }
                 cur.pos = p.rv.start;
                 cur.len = p.rv.end - p.rv.start + 1;
+                cur.infos = p.infos;
             }
             for(auto const & m : p.infos.getMemberNames())
             {

src/c++/lib/variant/VariantWriter.cpp

@@ -158,8 +158,8 @@ namespace variant
             int idx = bcf_hrec_find_key(hrec, "ID");
             if ( idx < 0 )
             {
-                // no ID => skip
-                bcf_hrec_destroy(hrec);
+                // no ID => pass on
+                bcf_hdr_add_hrec(hdr, hrec);
                 continue;
             }
 

src/python/Tools/vcfcallerinfo.py

@@ -45,8 +45,8 @@ def addVCF(self, vcfname):
         tf.close()
         vfh = {}
         try:
-            sp = subprocess.Popen("vcfhdr2json %s %s" % (vcfname, tf.name),
-                                  shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
+            sp = subprocess.Popen("vcfhdr2json '%s' '%s'" % (vcfname, tf.name),
+                shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
             o, e = sp.communicate()
 
             if sp.returncode != 0:
@@ -120,7 +120,7 @@ def addBAM(self, bamfile):
         """ Extract aligner information from a BAM file
         :param bamfile: name of BAM file
         """
-        sp = subprocess.Popen("samtools view -H %s" % bamfile,
+        sp = subprocess.Popen("samtools view -H '%s'" % bamfile,
                               shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         o, e = sp.communicate()
 

src/python/Tools/vcfextract.py

@@ -220,7 +220,7 @@ def extractHeadersJSON(vcfname):
     vfh = {}
 
     try:
-        sp = subprocess.Popen("vcfhdr2json %s %s" % (vcfname.replace(" ", "\\ "), tf.name.replace(" ", "\\ ")),
+        sp = subprocess.Popen("vcfhdr2json '%s' '%s'" % (vcfname, tf.name),
                               shell=True, stdout=subprocess.PIPE, stderr=subprocess.PIPE)
         o, e = sp.communicate()
 

src/python/hap.py

@@ -82,7 +82,7 @@ def main():
     parser.add_argument("--preprocess-truth", dest="preprocessing_truth", action="store_true", default=False,
                         help="Preprocess truth file with same settings as query (default is to accept truth in original format).")
     parser.add_argument("--usefiltered-truth", dest="usefiltered_truth", action="store_true", default=False,
-                        help="Preprocess truth file with same settings as query (default is to accept truth in original format).")
+                        help="Use filtered variant calls in truth file (by default, only PASS calls in the truth file are used)")
     parser.add_argument("--preprocessing-window-size", dest="preprocess_window",
                         default=10000, type=int,
                         help="Preprocessing window size (variants further apart than that size are not expected to interfere).")

src/python/pre.py

@@ -251,23 +251,23 @@ def updateArgs(parser):
                         default=False,
                         help="Left-shift variants safely.")
     parser.add_argument("--no-leftshift", dest="preprocessing_leftshift", action="store_false",
-                        help="Left-shift variants safely.")
+                        help="Do not left-shift variants safely.")
 
     parser.add_argument("--decompose", dest="preprocessing_decompose", action="store_true",
                         default=True,
                         help="Decompose variants into primitives. This results in more granular counts.")
     parser.add_argument("-D", "--no-decompose", dest="preprocessing_decompose", action="store_false",
-                        help="Decompose variants into primitives. This results in more granular counts.")
+                        help="Do not decompose variants into primitives.")
 
     parser.add_argument("--bcftools-norm", dest="preprocessing_norm", action="store_true", default=False,
                         help="Enable preprocessing through bcftools norm -c x -D (requires external "
                              " preprocessing to be switched on).")
 
     parser.add_argument("--fixchr", dest="fixchr", action="store_true", default=None,
-                        help="Add/remove chr prefix (default: auto, attempt to match reference).")
+                        help="Add chr prefix to VCF records where necessary (default: auto, attempt to match reference).")
 
     parser.add_argument("--no-fixchr", dest="fixchr", action="store_false",
-                        help="Add chr prefix to query file (default: auto, attempt to match reference).")
+                        help="Do not add chr prefix to VCF records (default: auto, attempt to match reference).")
 
     parser.add_argument("--bcf", dest="bcf", action="store_true", default=False,
                         help="Use BCF internally. This is the default when the input file"

src/sh/run_quantify_test.sh

@@ -52,9 +52,17 @@ if [[ $? != 0 ]]; then
 fi
 
 # these are all the metrics. We filter out all the command line bits
-cat ${TMP_OUT}.metrics.json | ${PYTHON} -mjson.tool | grep -v timestamp | grep -v hap.py > ${TMP_OUT}.hap.m.json
-cat ${TMP_OUT}.qfy.metrics.json | ${PYTHON} -mjson.tool | grep -v timestamp | grep -v qfy.py > ${TMP_OUT}.qfy.m.json
-diff ${TMP_OUT}.hap.m.json ${TMP_OUT}.qfy.m.json | head
+gunzip -c ${TMP_OUT}.metrics.json.gz | ${PYTHON} -mjson.tool | grep -v timestamp | grep -v hap.py > ${TMP_OUT}.hap.m.json
+if [[ $? != 0 ]] || [[ ! -s ${TMP_OUT}.hap.m.json ]]; then
+	echo "Cannot unzip metrics for original run."
+	exit 1
+fi
+gunzip -c ${TMP_OUT}.qfy.metrics.json.gz | ${PYTHON} -mjson.tool | grep -v timestamp | grep -v qfy.py > ${TMP_OUT}.qfy.m.json
+if [[ $? != 0 ]] || [[ ! -s ${TMP_OUT}.qfy.m.json ]]; then
+	echo "Cannot unzip metrics for re-quantified run."
+	exit 1
+fi
+diff ${TMP_OUT}.hap.m.json ${TMP_OUT}.qfy.m.json
 if [[ $? != 0 ]]; then
 	echo "Re-quantified counts are different! diff ${TMP_OUT}.hap.m.json ${TMP_OUT}.qfy.m.json "
 	exit 1