From 97a64db9ab33ef25d5f10f4a980dfc8c057bdf4d Mon Sep 17 00:00:00 2001
From: Jeremy Goecks <jeremy.goecks@emory.edu>
Date: Sun, 19 May 2013 11:59:50 -0400
Subject: [PATCH] Add library normalization and dispersion options to Cuffdiff
 wrapper.

---
 tools/ngs_rna/cuffdiff_wrapper.py  |  6 ++++++
 tools/ngs_rna/cuffdiff_wrapper.xml | 14 ++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/tools/ngs_rna/cuffdiff_wrapper.py b/tools/ngs_rna/cuffdiff_wrapper.py
index bbcefa3650d..4ba463df1b8 100644
--- a/tools/ngs_rna/cuffdiff_wrapper.py
+++ b/tools/ngs_rna/cuffdiff_wrapper.py
@@ -62,6 +62,8 @@ def __main__():
     parser.add_option( '-c', '--min-alignment-count', dest='min_alignment_count', help='The minimum number of alignments in a locus for needed to conduct significance testing on changes in that locus observed between samples. If no testing is performed, changes in the locus are deemed not signficant, and the locus\' observed changes don\'t contribute to correction for multiple testing. The default is 1,000 fragment alignments (up to 2,000 paired reads).' )
     parser.add_option( '--FDR', dest='FDR', help='The allowed false discovery rate. The default is 0.05.' )
     parser.add_option( '-u', '--multi-read-correct', dest='multi_read_correct', action="store_true", help='Tells Cufflinks to do an initial estimation procedure to more accurately weight reads mapping to multiple locations in the genome')
+    parser.add_option( '--library-norm-method', dest='library_norm_method' )
+    parser.add_option( '--dispersion-method', dest='dispersion_method' )
 
     # Advanced Options:	
     parser.add_option( '--num-importance-samples', dest='num_importance_samples', help='Sets the number of importance samples generated for each locus during abundance estimation. Default: 1000' )
@@ -143,6 +145,10 @@ def __main__():
     cmd = "cuffdiff --no-update-check -q"
     
     # Add options.
+    if options.library_norm_method:
+        cmd += ( " --library-norm-method %s" % options.library_norm_method )
+    if options.dispersion_method:
+        cmd += ( " --dispersion-method %s" % options.dispersion_method )
     if options.inner_dist_std_dev:
         cmd += ( " -s %i" % int ( options.inner_dist_std_dev ) )
     if options.num_threads:
diff --git a/tools/ngs_rna/cuffdiff_wrapper.xml b/tools/ngs_rna/cuffdiff_wrapper.xml
index e8cce23c66e..2be89aa3928 100644
--- a/tools/ngs_rna/cuffdiff_wrapper.xml
+++ b/tools/ngs_rna/cuffdiff_wrapper.xml
@@ -9,6 +9,8 @@
             --FDR=$fdr
             --num-threads="4"
             --min-alignment-count=$min_alignment_count
+            --library-norm-method=$library_norm_method
+            --dispersion-method=$dispersion_method
 
             --isoforms_fpkm_tracking_output=$isoforms_fpkm_tracking
             --genes_fpkm_tracking_output=$genes_fpkm_tracking
@@ -92,6 +94,18 @@
             </when>
         </conditional>
 
+        <param name="library_norm_method" type="select" label="Library normalization method">
+            <option value="geometric" selected="True">geometric</option>
+            <option value="classic-fpkm">classic-fpkm</option>
+            <option value="quartile">quartile</option>
+        </param>
+
+        <param name="dispersion_method" type="select" label="Dispersion estimation method" help="If using only one sample per condition, you must use 'blind.'">
+            <option value="pooled" selected="True">pooled</option>
+            <option value="per-condition">per-condition</option>
+            <option value="blind">blind</option>
+        </param>
+
         <param name="fdr" type="float" value="0.05" label="False Discovery Rate" help="The allowed false discovery rate."/>
 
         <param name="min_alignment_count" type="integer" value="10" label="Min Alignment Count" help="The minimum number of alignments in a locus for needed to conduct significance testing on changes in that locus observed between samples."/>