zzh
/
gatk-3.8
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

So, the VCF spec allows for an optional meta field in the header representing the date. However, using this field means that integration tests run on the vcf file will fail the MD5 test (which is what happened to the VariantFiltration test this morning after working just fine yesterday). 

After consulting our resident expert (Aaron), we're going to (temporarily) remove the date from the vcf output until we can come up with a better solution.  However, this shouldn't cause any short-term problems because the data truly is optional.
VF test's MD5s are updated.


git-svn-id: file:///humgen/gsa-scr1/gsa-engineering/svn_contents/trunk@1580 348d0f76-0448-11de-a6fe-93d51630548a
This commit is contained in:
ebanks 2009-09-10 14:28:43 +00:00
parent fc0d9578f6
commit e24c8d00d5
2 changed files with 15 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
java/src/org/broadinstitute/sting/playground/gatk/walkers/variantstovcf/VariantsToVCF.java
View File

@ -46,10 +46,12 @@ public class VariantsToVCF extends RefWalker<Integer, Integer> {
Map<String, String> metaData = new HashMap<String, String>();
List<String> additionalColumns = new ArrayList<String>();
Calendar cal = Calendar.getInstance();
// Don't output the data for now because it kills our unit test MD5s and is optional
// TODO - figure out what to do here
//Calendar cal = Calendar.getInstance();
//metaData.put("fileDate", String.format("%d%02d%02d", cal.get(Calendar.YEAR), cal.get(Calendar.MONTH), cal.get(Calendar.DAY_OF_MONTH)));
metaData.put("format", "VCRv3.2");
metaData.put("fileDate", String.format("%d%02d%02d", cal.get(Calendar.YEAR), cal.get(Calendar.MONTH), cal.get(Calendar.DAY_OF_MONTH)));
metaData.put("source", "VariantsToVCF");
metaData.put("reference", args.referenceFile.getAbsolutePath());

22
java/test/org/broadinstitute/sting/gatk/walkers/filters/VariantFiltrationIntegrationTest.java
View File

@ -9,63 +9,63 @@ public class VariantFiltrationIntegrationTest extends WalkerTest {
@Test
public void testIntervals() {
String[] md5DoC = {"c6ac824171bc2abcd6a896ba07d0c072", "21c8e1f9dc65fdfb39347547f9b04011"};
String[] md5DoC = {"492599c4d7d6dfca29659a7be3e3b7d4", "21c8e1f9dc65fdfb39347547f9b04011"};
WalkerTestSpec spec1 = new WalkerTestSpec(
"-T VariantFiltration -X DepthOfCoverage:max=70 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5DoC));
executeTest("testDoCFilter", spec1);
String[] md5AlleleBalance = {"06fa8901c6823766b89ec28acc435c47", "a13e4ce6260bf9f33ca99dc808b8e6ad"};
String[] md5AlleleBalance = {"33ce9e974efa7c095e3124f0ecad14b3", "a13e4ce6260bf9f33ca99dc808b8e6ad"};
WalkerTestSpec spec2 = new WalkerTestSpec(
"-T VariantFiltration -X AlleleBalance:low=0.25,high=0.75 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5AlleleBalance));
executeTest("testAlleleBalanceFilter", spec2);
String[] md5Strand = {"3f77d329a2f9223423a19665cf7eebbd", "0f7db0aad764268ee8fa3b857df8d87d"};
String[] md5Strand = {"f30ca865290e1c3537a8a30e4c3a5df2", "0f7db0aad764268ee8fa3b857df8d87d"};
WalkerTestSpec spec3 = new WalkerTestSpec(
"-T VariantFiltration -X FisherStrand:pvalue=0.0001 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5Strand));
executeTest("testStrandFilter", spec3);
String[] md5Lod = {"2c4251fc21998705458a9d25386f522f", "7e0c4f2b0fda85fd2891eee76c396a55"};
String[] md5Lod = {"948963861d9d2260e9f5ed6447aa30cb", "7e0c4f2b0fda85fd2891eee76c396a55"};
WalkerTestSpec spec4 = new WalkerTestSpec(
"-T VariantFiltration -X LodThreshold:lod=10 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5Lod));
executeTest("testLodFilter", spec4);
String[] md5MQ0 = {"8d0dbd951241177dddc4aab8b747ed86", "3203de335621851bccf596242b079e23"};
String[] md5MQ0 = {"57ac3b2df0590fe189e4462560cba686", "3203de335621851bccf596242b079e23"};
WalkerTestSpec spec5 = new WalkerTestSpec(
"-T VariantFiltration -X MappingQualityZero:max=70 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5MQ0));
executeTest("testMappingQuality0Filter", spec5);
String[] md5MQ = {"638719c44704fbac599dfc957a3dce06", "07c5e5b8b3b49a53a6617e63c9e56c71"};
String[] md5MQ = {"b8306b693dfee52113d8ecb405ddf25a", "ecc777feedea61f7b570d114c2ab89b1"};
WalkerTestSpec spec6 = new WalkerTestSpec(
"-T VariantFiltration -X MappingQuality:min=5 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
"-T VariantFiltration -X MappingQuality:min=20 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5MQ));
executeTest("testRMSMappingQualityFilter", spec6);
String[] md5OnOff = {"638719c44704fbac599dfc957a3dce06", "07c5e5b8b3b49a53a6617e63c9e56c71"};
String[] md5OnOff = {"8f35d20e9b53ed5c5a161b15501705bf", "67f2e1bc025833b0fa31f47195198997"};
WalkerTestSpec spec7 = new WalkerTestSpec(
"-T VariantFiltration -X OnOffGenotypeRatio:threshold=0.8 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
"-T VariantFiltration -X OnOffGenotypeRatio:threshold=0.9 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5OnOff));
executeTest("testOnOffGenotypeFilter", spec7);
String[] md5Clusters = {"22d6c46e3971573159d195c8d93e9293", "8fa6b6ffc93ee7fb8d6b52a7fb7815ef"};
String[] md5Clusters = {"f5c7c5da0198c4aaafdfcfb3c356eedc", "8fa6b6ffc93ee7fb8d6b52a7fb7815ef"};
WalkerTestSpec spec8 = new WalkerTestSpec(
"-T VariantFiltration -X ClusteredSnps:window=10,snps=3 -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,
Arrays.asList(md5Clusters));
executeTest("testClusteredSnpsFilter", spec8);
String[] md5Indels = {"3d9ebd5ada1c355e44245a64bc801b11", "8e0e915a1cb63d7049e0671ed00101fe"};
String[] md5Indels = {"30bf4c764f6dfd006d919ecaceee0166", "8e0e915a1cb63d7049e0671ed00101fe"};
WalkerTestSpec spec9 = new WalkerTestSpec(
"-T VariantFiltration -X IndelArtifact -B indels,PointIndel,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.indels -B cleaned,CleanedOutSNP,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.realigner_badsnps -R /broad/1KG/reference/human_b36_both.fasta -I /humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.1kg.p2.chr1_10mb_11_mb.SLX.bam -L 1:10,000,000-11,000,000 -B variant,Variants,/humgen/gsa-scr1/GATK_Data/Validation_Data/NA12878.chr1_10mb_11mb.slx.geli.calls -D /humgen/gsa-scr1/GATK_Data/dbsnp_129_b36.rod -vcf %s -included %s -sample NA12878",
2,