Merge branch 'release-0.7.0' into 'master'

Release 0.7.0



See merge request !449

.gitignore

@@ -10,7 +10,5 @@ git.properties
 .idea/*
 *.iml
 target/
-public/target/
-protected/target/
 site/
 *.sc
\ No newline at end of file

README.md

@@ -55,18 +55,18 @@ At the moment, we do not provide links to download the Biopet package. If you ar
 
 ## Contributing to Biopet
 
-Biopet is based on the Queue framework developed by the Broad Institute as part of their Genome Analysis Toolkit (GATK) framework. The current Biopet release is based on the GATK 3.4 release.
+Biopet is based on the Queue framework developed by the Broad Institute as part of their Genome Analysis Toolkit (GATK) framework. The current Biopet release is based on the GATK 3.5 release.
 
 We welcome any kind of contribution, be it merge requests on the code base, documentation updates, or any kinds of other fixes! The main language we use is Scala, though the repository also contains a small bit of Python and R. Our main code repository is located at [https://github.com/biopet/biopet](https://github.com/biopet/biopet/issues), along with our issue tracker.
 
 ## Local development setup
 
-To develop Biopet, Java 7, Maven 3.2.2, and GATK Queue 3.4 is required. Please consult the Java homepage and Maven homepage for the respective installation instruction. After you have both Java and Maven installed, you would then need to install GATK Queue. However, as the GATK Queue package is not yet available as an artifact in Maven Central, you will need to download, compile, and install GATK Queue first.
+To develop Biopet, Java 7, Maven 3.3.3, and GATK Queue 3.5 is required. Please consult the Java homepage and Maven homepage for the respective installation instruction. After you have both Java and Maven installed, you would then need to install GATK Queue. However, as the GATK Queue package is not yet available as an artifact in Maven Central, you will need to download, compile, and install GATK Queue first.
 
 ~~~
 $ git clone https://github.com/broadgsa/gatk-protected
-$ cd gatk-protected
-$ git checkout 3.4                              # the current release is based on GATK 3.4
+$ cd gatk
+$ git checkout 3.5                              # the current release is based on GATK 3.5
 $ mvn -U clean install
 ~~~
 

bam2wig/pom.xml

+<?xml version="1.0" encoding="UTF-8"?>
+<!--
+
+    Biopet is built on top of GATK Queue for building bioinformatic
+    pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+    SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+    should also be able to execute Biopet tools and pipelines.
+
+    Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+
+    Contact us at: sasc@lumc.nl
+
+    A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
+    license; For commercial users or users who do not want to follow the AGPL
+    license, please contact us to obtain a separate license.
+
+-->
+<project xmlns="http://maven.apache.org/POM/4.0.0"
+         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
+    <modelVersion>4.0.0</modelVersion>
+
+    <artifactId>Bam2Wig</artifactId>
+    <packaging>jar</packaging>
+
+    <parent>
+        <groupId>nl.lumc.sasc</groupId>
+        <artifactId>Biopet</artifactId>
+        <version>0.7.0</version>
+        <relativePath>../</relativePath>
+    </parent>
+
+    <inceptionYear>2014</inceptionYear>
+    <name>Bam2Wig</name>
+
+    <dependencies>
+        <dependency>
+            <groupId>nl.lumc.sasc</groupId>
+            <artifactId>BiopetCore</artifactId>
+            <version>${project.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>nl.lumc.sasc</groupId>
+            <artifactId>BiopetExtensions</artifactId>
+            <version>${project.version}</version>
+        </dependency>
+        <dependency>
+            <groupId>org.scalatest</groupId>
+            <artifactId>scalatest_2.10</artifactId>
+            <version>2.2.1</version>
+            <scope>test</scope>
+        </dependency>
+        <dependency>
+            <groupId>org.testng</groupId>
+            <artifactId>testng</artifactId>
+            <version>6.8</version>
+            <scope>test</scope>
+        </dependency>
+    </dependencies>
+
+</project>
\ No newline at end of file

public/bam2wig/src/main/scala/nl/lumc/sasc/biopet/pipelines/bamtobigwig/Bam2Wig.scala → bam2wig/src/main/scala/nl/lumc/sasc/biopet/pipelines/bamtobigwig/Bam2Wig.scala

@@ -8,8 +8,7 @@
  *
  * Contact us at: sasc@lumc.nl
  *
- * A dual licensing mode is applied. The source code within this project that are
- * not part of GATK Queue is freely available for non-commercial use under an AGPL
+ * A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
  * license; For commercial users or users who do not want to follow the AGPL
  * license, please contact us to obtain a separate license.
  */

public/bam2wig/src/main/scala/nl/lumc/sasc/biopet/pipelines/bamtobigwig/BamToChromSizes.scala → bam2wig/src/main/scala/nl/lumc/sasc/biopet/pipelines/bamtobigwig/BamToChromSizes.scala

@@ -8,8 +8,7 @@
  *
  * Contact us at: sasc@lumc.nl
  *
- * A dual licensing mode is applied. The source code within this project that are
- * not part of GATK Queue is freely available for non-commercial use under an AGPL
+ * A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
  * license; For commercial users or users who do not want to follow the AGPL
  * license, please contact us to obtain a separate license.
  */

public/bam2wig/src/test/resources/log4j.properties → bam2wig/src/test/resources/log4j.properties

@@ -8,8 +8,7 @@
 #
 # Contact us at: sasc@lumc.nl
 #
-# A dual licensing mode is applied. The source code within this project that are
-# not part of GATK Queue is freely available for non-commercial use under an AGPL
+# A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
 # license; For commercial users or users who do not want to follow the AGPL
 # license, please contact us to obtain a separate license.
 #

bam2wig/src/test/scala/nl/lumc/sasc/biopet/pipelines/bamtobigwig/BamToChromSizesTest.scala

+/**
+ * Biopet is built on top of GATK Queue for building bioinformatic
+ * pipelines. It is mainly intended to support LUMC SHARK cluster which is running
+ * SGE. But other types of HPC that are supported by GATK Queue (such as PBS)
+ * should also be able to execute Biopet tools and pipelines.
+ *
+ * Copyright 2014 Sequencing Analysis Support Core - Leiden University Medical Center
+ *
+ * Contact us at: sasc@lumc.nl
+ *
+ * A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
+ * license; For commercial users or users who do not want to follow the AGPL
+ * license, please contact us to obtain a separate license.
+ */
+package nl.lumc.sasc.biopet.pipelines.bamtobigwig
+
+import java.io.File
+import java.nio.file.Paths
+
+import org.scalatest.Matchers
+import org.scalatest.testng.TestNGSuite
+import org.testng.annotations.Test
+
+import scala.io.Source
+
+/**
+ * Created by pjvanthof on 09/05/16.
+ */
+class BamToChromSizesTest extends TestNGSuite with Matchers {
+  private def resourcePath(p: String): String = {
+    Paths.get(getClass.getResource(p).toURI).toString
+  }
+
+  @Test
+  def testChromSizes: Unit = {
+    val bamFile = new File(resourcePath("/empty.bam"))
+    val bamToChromSizes = new BamToChromSizes(null)
+    bamToChromSizes.bamFile = bamFile
+    bamToChromSizes.chromSizesFile = File.createTempFile("chrom.", ".sizes")
+    bamToChromSizes.chromSizesFile.deleteOnExit()
+    bamToChromSizes.run()
+    Source.fromFile(bamToChromSizes.chromSizesFile).getLines().toList shouldBe List("chrQ\t10000", "chrR\t10000")
+  }
+}

public/bammetrics/pom.xml → bammetrics/pom.xml

@@ -9,8 +9,7 @@
 
     Contact us at: sasc@lumc.nl
 
-    A dual licensing mode is applied. The source code within this project that are
-    not part of GATK Queue is freely available for non-commercial use under an AGPL
+    A dual licensing mode is applied. The source code within this project is freely available for non-commercial use under an AGPL
     license; For commercial users or users who do not want to follow the AGPL
     license, please contact us to obtain a separate license.
 
@@ -25,7 +24,7 @@
     <parent>
         <groupId>nl.lumc.sasc</groupId>
         <artifactId>Biopet</artifactId>
-        <version>0.6.0</version>
+        <version>0.7.0</version>
         <relativePath>../</relativePath>
     </parent>
     

public/bammetrics/src/main/resources/nl/lumc/sasc/biopet/pipelines/bammetrics/alignmentSummary.ssp → bammetrics/src/main/resources/nl/lumc/sasc/biopet/pipelines/bammetrics/alignmentSummary.ssp

@@ -63,6 +63,7 @@
         #if (!sampleLevel) <th>Library</th> #end
         <th>Total</th>
         <th>Mapped</th>
+        <th>Secondary</th>
         <th>(%)</th>
         <th>Duplicates</th>
         <th>(%)</th>
@@ -85,12 +86,14 @@
                         val total = summary.getValue((prefixPath ::: List("biopet_flagstat", "All")):_*).getOrElse(0L).asInstanceOf[Long]
                         val mapped = summary.getValue((prefixPath ::: List("biopet_flagstat", "Mapped")):_*).getOrElse(0L).asInstanceOf[Long]
                         val duplicates = summary.getValue((prefixPath ::: List("biopet_flagstat", "Duplicates")):_*).getOrElse(0L).asInstanceOf[Long]
+                        val secondary = summary.getValue((prefixPath ::: List("biopet_flagstat", "NotPrimaryAlignment")):_*).getOrElse(0L).asInstanceOf[Long]
                     }#
                     <td>${total}</td>
                     <td>${mapped}</td>
-                    <td>${mapped.toDouble / total * 100}%</td>
+                    <td>${secondary}</td>
+                    <td>${(mapped - secondary).toDouble / (total - secondary) * 100}%</td>
                     <td>${duplicates}</td>
-                    <td>${duplicates.toDouble / total * 100}%</td>
+                    <td>${duplicates.toDouble / (total - secondary) * 100}%</td>
                 </tr>
             #end
         #end