Merge pull request #106 from PolinaBevad/chimeric+unique

Added chimeric and unique modes from Perl version

Author: BioInfo

Readme.md

@@ -242,8 +242,8 @@ The VarDictJava program follows the workflow:
     The minimum # of reads to determine strand bias, default: `2`
 - `-Q INT`  
     If set, reads with mapping quality less than INT will be filtered and ignored
-- `-q INT`   
-    The phred score for a base to be considered a good call.  Default: 25 (for Illumina). For PGM, set it to ~15, as PGM tends to underestimate base quality.
+- `-q double`   
+    The phred score for a base to be considered a good call.  Default: 22.5 (for Illumina). For PGM, set it to ~15, as PGM tends to underestimate base quality.
 - `-m INT`   
     If set, reads with mismatches more than `INT` will be filtered and ignored.  Gaps are not counted as mismatches. Valid only for bowtie2/TopHat or BWA aln followed by sampe.  BWA mem is calculated as NM - Indels.  Default: 8, or reads with more than 8 mismatches will not be used.
 - `-T INT`  
@@ -274,7 +274,13 @@ The VarDictJava program follows the workflow:
      `LENIENT`  - Emit warnings but keep going if possible.
      `SILENT`   - Like `LENIENT`, only don't emit warning messages.
     Default: `LENIENT`
-
+- `-u`  
+    Indicate unique mode, which when mate pairs overlap, the overlapping part will be counted only once using forward read only.
+    Default: unique mode disabled, all reads are counted.
+- `--chimeric`
+    Indicate to turn off chimeric reads filtering.  Chimeric reads are artifacts from library construction, 
+    where a read can be split into two segments, each will be aligned within 1-2 read length distance,
+    but in opposite direction. Default: filtering enabled
 ## Output columns
 
 1. Sample - sample name

src/main/java/com/astrazeneca/vardict/Configuration.java

@@ -67,7 +67,7 @@ public class Configuration {
     /**
      * The phred score for a base to be considered a good call
      */
-    int goodq; // -q, default = 23
+    double goodq; // -q, default = 22.5
     final int buffer = 200;
     /**
      * Extension of bp to look for mismatches after insersion or deletion
@@ -143,12 +143,24 @@ public class Configuration {
      */
     boolean includeNInTotalDepth = false; // -K
 
+    /**
+     * Indicate unique mode, which when mate pairs overlap,
+     * the overlapping part will be counted only once using forward read only.
+     */
+    boolean uniqueModeOn = false; // -u
 
     /**
      * Threads count
      */
     int threads;
 
+    /**
+     * Indicate to turn off chimeric reads filtering.  Chimeric reads are artifacts from library construction,
+     * where a read can be split into two segments, each will be aligned within 1-2 read length distance,
+     * but in opposite direction.
+     */
+    public boolean chimeric = false; // --chimeric
+
     public boolean isColumnForChromosomeSet() {
         return columnForChromosome >= 0;
     }

src/main/java/com/astrazeneca/vardict/Main.java

@@ -93,13 +93,13 @@ private void run(CommandLine cmd) throws ParseException, IOException {
         conf.columnForChromosome = getColumnValue(cmd, "c", -1);
 
         conf.numberNucleotideToExtend = getIntValue(cmd, "x", 0);
-        conf.freq = getDoubleValue(cmd, "f", 0.05d);
+        conf.freq = getDoubleValue(cmd, "f", 0.01d);
         conf.minr = getIntValue(cmd, "r", 2);
         conf.minb = getIntValue(cmd, "B", 2);
         if (cmd.hasOption("Q")) {
             conf.mappingQuality = ((Number)cmd.getParsedOptionValue("Q")).intValue();
         }
-        conf.goodq = getIntValue(cmd, "q", 25);
+        conf.goodq = getDoubleValue(cmd, "q", 22.5);
         conf.mismatch =  getIntValue(cmd, "m", 8);
         conf.trimBasesAfter = getIntValue(cmd, "T", 0);
         conf.vext = getIntValue(cmd, "X", 3);
@@ -133,6 +133,15 @@ private void run(CommandLine cmd) throws ParseException, IOException {
             conf.includeNInTotalDepth = true;
         }
 
+        if (cmd.hasOption("chimeric")) {
+            conf.chimeric = true;
+        }
+
+
+        if (cmd.hasOption("u")) {
+            conf.uniqueModeOn = true;
+        }
+
         conf.threads = Math.max(readThreadsCount(cmd), 1);
 
         VarDict.start(conf);
@@ -181,6 +190,8 @@ private static Options buildOptions() {
         options.addOption("t", false, "Indicate to remove duplicated reads.  Only one pair with same start positions will be kept");
         options.addOption("3", false, "Indicate to move indels to 3-prime if alternative alignment can be achieved.");
         options.addOption("K", false, "Include Ns in the total depth calculation");
+        options.addOption("u", false, "Indicate unique mode, which when mate pairs overlap, the overlapping part will be counted only once using forward read only.");
+        options.addOption("chimeric", false, "Indicate to turn off chimeric reads filtering.");
 
         options.addOption(OptionBuilder.withArgName("bit")
                 .hasArg(true)