# HG changeset patch
# User rnateam
# Date 1404839070 14400
# Node ID f973ec6e5192ab06f5a12696ce1f3d0874e11e60
# Parent  a92d8689aa30be7572d476ebac5f8c0b254b7da9
Uploaded

diff -r a92d8689aa30 -r f973ec6e5192 blockclust.xml
--- a/blockclust.xml	Tue Jul 08 11:59:49 2014 -0400
+++ b/blockclust.xml	Tue Jul 08 13:04:30 2014 -0400
@@ -1,5 +1,5 @@
 <tool id="blockclust" name="BlockClust" version="1.0.0">
-    <description>Non-coding RNA clustering from deep sequencing read profiles</description>
+    <description>efficient clustering and classification of non-coding RNAs from short read RNA-seq profiles</description>
     <requirements>
         <requirement type="package" version="1.0">blockclust</requirement>
         <requirement type="package" version="1.1">eden</requirement>
@@ -10,7 +10,6 @@
     </requirements>
     <version_command>echo '1.0'</version_command>
     <command>
-
         #if str($tool_mode.operation) == "pre":
             BlockClustPipeLine.pl -m PRE -bam $tool_mode.reads_bam -tbed $tags_bed
         #elif str($tool_mode.operation) == "clust":
@@ -73,7 +72,7 @@
         <conditional name="tool_mode">
             <param name="operation" type="select" label="Select mode of operation">
                 <option value="pre">Pre-processing </option>
-                <option value="clust">Clustering</option>
+                <option value="clust">Clustering and classification</option>
                 <option value="post">Post-processing</option>
             </param>
             <when value="pre">
@@ -119,10 +118,10 @@
         <data format="pdf" name="hclust_plot" label="BlockClust: Hierarchical clustering plot on ${on_string}" >
             <filter> tool_mode["operation"]=="clust"</filter>
         </data>
-        <data format="bed" name="clusters" label="BlockClust: Clustering BED on ${on_string}">
+        <data format="bed" name="clusters" label="BlockClust: BED of predicted clusters on ${on_string}">
             <filter> tool_mode["operation"]=="clust"</filter>
         </data>
-        <data format="bed" name="model_based_pred_bed" label="BlockClust: Model based prediction BED on ${on_string}">
+        <data format="bed" name="model_based_pred_bed" label="BlockClust: Model based predictions BED on ${on_string}">
             <filter>
             ((
                 tool_mode["operation"] == 'clust' and 
@@ -131,7 +130,7 @@
              ))
              </filter>
         </data>
-        <data format="bed" name="nearest_neighbour_pred_bed" label="BlockClust: Nearest neighbour prediction BED on ${on_string}">
+        <data format="bed" name="nearest_neighbour_pred_bed" label="BlockClust: Nearest neighbor predictions BED on ${on_string}">
             <filter>
             ((
                 tool_mode["operation"] == 'clust' and 
@@ -156,26 +155,57 @@
 
 **What it does** 
 
-Clusters the read profiles (i.e., blockgroups) from the blockbuster tool.
+BlockClust is an efficient approach to detect transcripts with similar 
+processing patterns. We propose a novel way to encode expression profiles
+in compact discrete structures, which can then be processed using 
+fast graph-kernel techniques. BlockClust allows both clustering and 
+classification of small non-coding RNAs.
+
+BlockClust runs in three modes: 
+1) Pre-processing - converts given mapped reads (BAM) into BED file of tags
+2) Clustering and classification - of given input block groups (from blockbuster tool) as explained in [1]_
+3) Post-processing - extracts distribution of clusters searched against Rfam database and plots hierarchical clustering made out of centroids of each BlockClust predicted cluster.
+
+For a thorough analysis of your data, we suggest you to use complete blockclust workflow, which contains all three modes of operation.
 
 **Inputs**
 
-BlockClust needs output of tool blockbuster as input 
+BlockClust input files are dependent on the mode of operation:
+1) Pre-processing mode:
+Binary Sequence Alignment Map (BAM) file
+
+2) Clustering and classification:
+A blockgroups file generated by blockbuster tool
+Select reference genome
+
+3) Post-processing:
+Output of cmsearch, searched clusters generated by BlockClust against Rfam
+BED file containing clusters generated by BlockClust
+Pairwise similarities of blockgroups generated by BlockClust
 
 **Output**
+1) Pre-processing mode:
+BED file of tags with expressions
 
-BlockClust produces a fasta file containing clusters.
+2) Clustering and classification:
+Hierarchical clustering plot of all input blockgroups by their similarity
+Pairwise similarities of all input blockgroups
+BED file containing predicted clusters
+BED file containing prediction of blockgroups by pre-compiled SVM binary classification model.
+
+3) Post-processing:
+Distribution of clusters with annotations searched against Rfam database
+hierarchical clustering made out of centroids of each BlockClust predicted cluster
 
 ------
 
 **Licenses**
 
-If **BlockClust** is used to obtain results for scientific publications it
-should be cited as [1]_.
+If **BlockClust** is used to obtain results for scientific publications it should be cited as [1]_.
 
 **References** 
 
-.. [1] 
+[1] Pavankumar Videm, Dominic Rose, Fabrizio Costa, and Rolf Backofen. "BlockClust: efficient clustering and classification of non-coding RNAs from short read RNA-seq profiles." Bioinformatics 30, no. 12 (2014): i274-i282.
 
 
     </help>