Repository 'data_manager_salmon_index_builder'
hg clone https://toolshed.g2.bx.psu.edu/repos/ieguinoa/data_manager_salmon_index_builder

Changeset 1:c5dea2080109 (2018-08-14)
Previous changeset 0:6cd60ba8a842 (2018-08-14) Next changeset 2:f7d9182bdcab (2018-10-19)
Commit message:
Uploaded
modified:
data_manager_conf.xml
tool_data_table_conf.xml.sample
added:
.shed.yml
data_manager/salmon_index_builder.py
data_manager/salmon_index_builder.xml
tool-data/all_fasta.loc.sample
tool-data/salmon_indexes.loc.sample
removed:
README.md
data_manager/data_manager_fetch_gff.py
data_manager/data_manager_fetch_gff.xml
tool-data/all_gff.loc.sample
tool-data/representative_gff.loc.sample
b
diff -r 6cd60ba8a842 -r c5dea2080109 .shed.yml
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/.shed.yml Tue Aug 14 11:19:49 2018 -0400
b
@@ -0,0 +1,19 @@
+categories:
+- Data Managers
+description: Pre-generate indexes for Salmon
+homepage_url: https://github.com/COMBINE-lab/salmon
+long_description: |
+  Salmon is a wicked-fast program to produce a highly-accurate, 
+  transcript-level quantification estimates from RNA-seq data. 
+  Salmon achieves is accuracy and speed via a number of different innovations, 
+  including the use of quasi-mapping (accurate but fast-to-compute proxies for traditional read alignments), 
+  and massively-parallel stochastic collapsed variational inference. 
+  The result is a versatile tool that fits nicely into many differnt pipelines. 
+  For example, you can choose to make use of our quasi-mapping algorithm by providing 
+  Salmon with raw sequencing reads, or, if it is more convenient, you can provide 
+  Salmon with regular alignments (e.g. an unsorted BAM file produced with your favorite aligner), 
+  and it will use the same wicked-fast, state-of-the-art inference algorithm to estimate transcript-level abundances for your experiment.
+name: data_manager_salmon_index_builder
+owner: iuc
+remote_repository_url: https://github.com/ieguinoa/data_manager_salmon_index_builder
+type: unrestricted
b
diff -r 6cd60ba8a842 -r c5dea2080109 README.md
--- a/README.md Tue Aug 14 11:14:52 2018 -0400
+++ /dev/null Thu Jan 01 00:00:00 1970 +0000
b
@@ -1,2 +0,0 @@
-# data_manager_fetch_gff
-Galaxy Data Manager to fetch gene annotation files
b
diff -r 6cd60ba8a842 -r c5dea2080109 data_manager/data_manager_fetch_gff.py
--- a/data_manager/data_manager_fetch_gff.py Tue Aug 14 11:14:52 2018 -0400
+++ /dev/null Thu Jan 01 00:00:00 1970 +0000
[
b'@@ -1,445 +0,0 @@\n-#!/usr/bin/env python\n-#Dan Blankenberg\n-\n-import sys\n-import os\n-import tempfile\n-import shutil\n-import optparse\n-from ftplib import FTP\n-import tarfile\n-import zipfile\n-import gzip\n-import bz2\n-try:\n-    # For Python 3.0 and later\n-    from urllib.request import urlopen\n-    from io import BytesIO as StringIO\n-    from io import UnsupportedOperation\n-except ImportError:\n-    # Fall back to Python 2\'s urllib2\n-    from urllib2 import urlopen\n-    from StringIO import StringIO\n-    UnsupportedOperation = AttributeError\n-from json import loads, dumps\n-\n-\n-CHUNK_SIZE = 2**20  # 1mb\n-\n-DATA_TABLE_NAME = \'all_gff\'\n-\n-def cleanup_before_exit( tmp_dir ):\n-    if tmp_dir and os.path.exists( tmp_dir ):\n-        shutil.rmtree( tmp_dir )\n-\n-\n-def stop_err(msg):\n-    sys.stderr.write(msg)\n-    sys.exit(1)\n-\n-\n-def get_dbkey_dbname_id_name( params, dbkey_description=None ):\n-#    dbkey = params[\'param_dict\'][\'dbkey_source\'][\'dbkey\']\n-    #TODO: ensure sequence_id is unique and does not already appear in location file\n-    sequence_id = params[\'param_dict\'][\'sequence_id\']\n-    if not sequence_id:\n-        sequence_id = dbkey #uuid.uuid4() generate and use an uuid instead?\n-    \n-#    if params[\'param_dict\'][\'dbkey_source\'][\'dbkey_source_selector\'] == \'new\':\n-#        dbkey_name = params[\'param_dict\'][\'dbkey_source\'][\'dbkey_name\']\n-#        if not dbkey_name:\n-#            dbkey_name = dbkey\n-#    else:\n-#        dbkey_name = None\n-    dbkey = params[\'param_dict\'][\'dbkey\'] \n-    dbkey_name = dbkey_description\n-    sequence_name = params[\'param_dict\'][\'sequence_name\']\n-    if not sequence_name:\n-        sequence_name = dbkey_description\n-        if not sequence_name:\n-            sequence_name = dbkey\n-    return dbkey, dbkey_name, sequence_id, sequence_name\n-\n-\n-def _get_files_in_ftp_path( ftp, path ):\n-    path_contents = []\n-    ftp.retrlines( \'MLSD %s\' % ( path ), path_contents.append )\n-    return [ line.split( \';\' )[ -1 ].lstrip() for line in path_contents ]\n-\n-\n-def _get_stream_readers_for_tar( fh, tmp_dir ):\n-    fasta_tar = tarfile.open( fileobj=fh, mode=\'r:*\' )\n-    return [x for x in [fasta_tar.extractfile(member) for member in fasta_tar.getmembers()] if x]\n-\n-\n-def _get_stream_readers_for_zip( fh, tmp_dir ):\n-    """\n-    Unpacks all archived files in a zip file.\n-    Individual files will be concatenated (in _stream_fasta_to_file)\n-    """\n-    fasta_zip = zipfile.ZipFile( fh, \'r\' )\n-    rval = []\n-    for member in fasta_zip.namelist():\n-        fasta_zip.extract( member, tmp_dir )\n-        rval.append( open( os.path.join( tmp_dir, member ), \'rb\' ) )\n-    return rval\n-\n-\n-def _get_stream_readers_for_gzip( fh, tmp_dir ):\n-    return [ gzip.GzipFile( fileobj=fh, mode=\'rb\') ]\n-\n-\n-def _get_stream_readers_for_bz2( fh, tmp_dir ):\n-    return [ bz2.BZ2File( fh.name, \'rb\') ]\n-\n-\n-def sort_fasta( fasta_filename, sort_method, params ):\n-    if sort_method is None:\n-        return\n-    assert sort_method in SORTING_METHODS, ValueError( "%s is not a valid sorting option." % sort_method )\n-    return SORTING_METHODS[ sort_method ]( fasta_filename, params )\n-\n-\n-def _move_and_index_fasta_for_sorting( fasta_filename ):\n-    unsorted_filename = tempfile.NamedTemporaryFile().name\n-    shutil.move( fasta_filename, unsorted_filename )\n-    fasta_offsets = {}\n-    unsorted_fh = open( unsorted_filename )\n-    while True:\n-        offset = unsorted_fh.tell()\n-        line = unsorted_fh.readline()\n-        if not line:\n-            break\n-        if line.startswith( ">" ):\n-            line = line.split( None, 1 )[0][1:]\n-            fasta_offsets[ line ] = offset\n-    unsorted_fh.close()\n-    current_order = map( lambda x: x[1], sorted( map( lambda x: ( x[1], x[0] ), fasta_offsets.items() ) ) )\n-    return ( unsorted_filename, fasta_offsets, current_order )\n-\n-\n-def _write_sorted_fasta( sorted_names, fasta_offsets, sorted_fasta_filename, unsorted_fasta_filename ):\n-    unsorted_fh = open( unsorted_fasta_filename )\n-    sorted_fh'..b'fh.read( CHUNK_SIZE )\n-                    if data:\n-                        fasta_writer.write( data )\n-                        last_char = data[-1]\n-                    else:\n-                        break\n-                if close_stream:\n-                    fh.close()\n-        else:\n-            while True:\n-                data = fasta_stream.read( CHUNK_SIZE )\n-                if data:\n-                    fasta_writer.write( data )\n-                else:\n-                    break\n-            if close_stream:\n-                fasta_stream.close()\n-\n-    #sort_fasta( fasta_filename, params[\'param_dict\'][\'sorting\'][\'sort_selector\'], params )\n-    \n-    \n-    return [ ( DATA_TABLE_NAME, dict( value=sequence_id, dbkey=dbkey, name=sequence_name, path=fasta_base_filename ) ) ]\n-\n-\n-def compute_fasta_length( fasta_file, out_file, keep_first_word=False ):\n-\n-    infile = fasta_file\n-    out = open( out_file, \'w\')\n-\n-    fasta_title = \'\'\n-    seq_len = 0\n-\n-    first_entry = True\n-\n-    for line in open( infile ):\n-        line = line.strip()\n-        if not line or line.startswith( \'#\' ):\n-            continue\n-        if line[0] == \'>\':\n-            if first_entry == False:\n-                if keep_first_word:\n-                    fasta_title = fasta_title.split()[0]\n-                out.write( "%s\\t%d\\n" % ( fasta_title[ 1: ], seq_len ) )\n-            else:\n-                first_entry = False\n-            fasta_title = line\n-            seq_len = 0\n-        else:\n-            seq_len += len(line)\n-\n-    # last fasta-entry\n-    if keep_first_word:\n-        fasta_title = fasta_title.split()[0]\n-    out.write( "%s\\t%d\\n" % ( fasta_title[ 1: ], seq_len ) )\n-    out.close()\n-\n-\n-def _create_symlink( input_filename, target_directory, dbkey, dbkey_name, sequence_id, sequence_name ):\n-    fasta_base_filename = "%s.fa" % sequence_id\n-    fasta_filename = os.path.join( target_directory, fasta_base_filename )\n-    os.symlink( input_filename, fasta_filename )\n-    return [  ( DATA_TABLE_NAME, dict( value=sequence_id, dbkey=dbkey, name=sequence_name, path=fasta_base_filename ) ) ]\n-\n-\n-REFERENCE_SOURCE_TO_DOWNLOAD = dict( ucsc=download_from_ucsc, ncbi=download_from_ncbi, url=download_from_url, history=download_from_history, directory=copy_from_directory )\n-\n-SORTING_METHODS = dict( as_is=_sort_fasta_as_is, lexicographical=_sort_fasta_lexicographical, gatk=_sort_fasta_gatk, custom=_sort_fasta_custom )\n-\n-\n-def main():\n-    #Parse Command Line\n-    parser = optparse.OptionParser()\n-    parser.add_option( \'-d\', \'--dbkey_description\', dest=\'dbkey_description\', action=\'store\', type="string", default=None, help=\'dbkey_description\' )\n-    parser.add_option( \'-t\', \'--type\', dest=\'file_type\', action=\'store\', type=\'string\', default=None, help=\'file_type\')\n-    (options, args) = parser.parse_args()\n-    \n-    filename = args[0]\n-    global DATA_TABLE_NAME\n-    if options.file_type == \'representative\':\n-       DATA_TABLE_NAME= \'representative_gff\'\n-    params = loads( open( filename ).read() )\n-    target_directory = params[ \'output_data\' ][0][\'extra_files_path\']\n-    os.mkdir( target_directory )\n-    data_manager_dict = {}\n-    \n-    dbkey, dbkey_name, sequence_id, sequence_name = get_dbkey_dbname_id_name( params, dbkey_description=options.dbkey_description ) \n-    \n-    if dbkey in [ None, \'\', \'?\' ]:\n-        raise Exception( \'"%s" is not a valid dbkey. You must specify a valid dbkey.\' % ( dbkey ) )\n-\n-    # Create a tmp_dir, in case a zip file needs to be uncompressed\n-    tmp_dir = tempfile.mkdtemp()\n-    #Fetch the FASTA\n-    try:\n-        REFERENCE_SOURCE_TO_DOWNLOAD[ params[\'param_dict\'][\'reference_source\'][\'reference_source_selector\'] ]( data_manager_dict, params, target_directory, dbkey, dbkey_name, sequence_id, sequence_name, tmp_dir )\n-    finally:\n-        cleanup_before_exit(tmp_dir)\n-    #save info to json file\n-    open( filename, \'wb\' ).write( dumps( data_manager_dict ).encode() )\n-        \n-if __name__ == "__main__":\n-    main()\n'
b
diff -r 6cd60ba8a842 -r c5dea2080109 data_manager/data_manager_fetch_gff.xml
--- a/data_manager/data_manager_fetch_gff.xml Tue Aug 14 11:14:52 2018 -0400
+++ /dev/null Thu Jan 01 00:00:00 1970 +0000
[
@@ -1,66 +0,0 @@
-<tool id="data_manager_fetch_gff" name="Create entries in gff data table" version="0.0.1" tool_type="manage_data">
-    <description>fetching</description>
-    <command><![CDATA[
-       python "$__tool_directory__"/data_manager_fetch_gff.py "${out_file}"
-       --type $file_type
-       --dbkey_description ${ dbkey.get_display_text() }
-        
-    ]]></command>
-    <inputs>
-         <param name="file_type" type="select" label="GFF file with only one representative transcript per gene (for htseq-count use) or full features file">
-                <option value="representative">Representative GFF</option>
-                <option value="full">GFF with complete features</option>
-            </param>

-        <param name="dbkey" type="genomebuild" label="DBKEY to assign to data" />
-        <param type="text" name="sequence_name" value="" label="Name of sequence" />
-        <param type="text" name="sequence_id" value="" label="ID for sequence" />
-        <conditional name="reference_source">
-            <param name="reference_source_selector" type="select" label="Choose the source for the reference genome">
-                <option value="url">URL</option>
-                <option value="history">History</option>
-                <option value="directory">Directory on Server</option>
-            </param>
-            <when value="url">
-                <param type="text" area="True" name="user_url" value="http://" label="URLs" optional="False" />
-            </when>
-            <when value="history">
-                <param name="input_fasta" type="data" format="fasta" label="FASTA File" multiple="False" optional="False" />
-            </when>
-            <when value="directory">
-                <param type="text" name="fasta_filename" value="" label="Full path to FASTA File on disk" optional="False" />
-                <param type="boolean" name="create_symlink" truevalue="create_symlink" falsevalue="copy_file" label="Create symlink to original data instead of copying" checked="False" />
-            </when>
-        </conditional>
-    </inputs>
-    <outputs>
-        <data name="out_file" format="data_manager_json"/>
-    </outputs>
-    <tests>
-        <!-- TODO: need some way to test that new entry was added to data table -->
-        <test>
-            <param name="dbkey" value="anoGam1"/>
-            <param name="sequence_name" value=""/>
-            <param name="sequence_desc" value=""/>
-            <param name="sequence_id" value=""/>
-            <param name="reference_source_selector" value="history"/>
-            <param name="input_fasta" value="phiX174.fasta"/>
-            <param name="sort_selector" value="as_is"/>
-            <output name="out_file" file="phiX174.data_manager_json"/>
-        </test>
-    </tests>
-    <help>
-**What it does**
-
-Fetches a gff file from various sources (URL, Galaxy History, or a server directory) and populates the "all_gff" data table.
-
-------
-
-
-
-.. class:: infomark
-
-**Notice:** If you leave name, description, or id blank, it will be generated automatically.
-
-    </help>
-</tool>
b
diff -r 6cd60ba8a842 -r c5dea2080109 data_manager/salmon_index_builder.py
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/data_manager/salmon_index_builder.py Tue Aug 14 11:19:49 2018 -0400
[
@@ -0,0 +1,82 @@
+#!/usr/bin/env python
+# Based heavily on the kallisto data manager wrapper script by iuc
+from __future__ import print_function
+
+import argparse
+import os
+import subprocess
+import sys
+from json import dumps, loads
+
+DEFAULT_DATA_TABLE_NAME = "salmon_indexes"
+
+
+def get_id_name( params, dbkey, fasta_description=None):
+    # TODO: ensure sequence_id is unique and does not already appear in location file
+    sequence_id = params['param_dict']['sequence_id']
+    if not sequence_id:
+        sequence_id = dbkey
+
+    sequence_name = params['param_dict']['sequence_name']
+    if not sequence_name:
+        sequence_name = fasta_description
+        if not sequence_name:
+            sequence_name = dbkey
+    return sequence_id, sequence_name
+
+
+def build_salmon_index( data_manager_dict, options, params, sequence_id, sequence_name ):
+    data_table_name = options.data_table_name or DEFAULT_DATA_TABLE_NAME
+    target_directory = params[ 'output_data' ][0]['extra_files_path']
+    if not os.path.exists( target_directory ):
+        os.mkdir( target_directory )
+    if options.kmer_size != '':
+        args.append('-k')
+        args.append(options.kmer_size)
+    args.extend( [ '-t' , options.fasta_filename, '-i', sequence_id ] )
+    proc = subprocess.Popen( args=args, shell=False, cwd=target_directory )
+    return_code = proc.wait()
+    if return_code:
+        print("Error building index.", file=sys.stderr)
+        sys.exit( return_code )
+    data_table_entry = dict( value=sequence_id, dbkey=options.fasta_dbkey, name=sequence_name, path=sequence_id )
+    _add_data_table_entry( data_manager_dict, data_table_name, data_table_entry )
+
+
+def _add_data_table_entry( data_manager_dict, data_table_name, data_table_entry ):
+    data_manager_dict['data_tables'] = data_manager_dict.get( 'data_tables', {} )
+    data_manager_dict['data_tables'][ data_table_name ] = data_manager_dict['data_tables'].get( data_table_name, [] )
+    data_manager_dict['data_tables'][ data_table_name ].append( data_table_entry )
+    return data_manager_dict
+
+
+def main():
+    # Parse Command Line
+    parser = argparse.ArgumentParser()
+    parser.add_argument( '--output', dest='output', action='store', type=str, default=None )
+    parser.add_argument( '--fasta_filename', dest='fasta_filename', action='store', type=str, default=None )
+    parser.add_argument( '--fasta_dbkey', dest='fasta_dbkey', action='store', type=str, default=None )
+    parser.add_argument( '--fasta_description', dest='fasta_description', action='store', type=str, default=None )
+    parser.add_argument( '--data_table_name', dest='data_table_name', action='store', type=str, default='salmon_indexes' )
+    parser.add_argument( '-k', '--kmer_size', dest='kmer_size', action='store', type=str, help='kmer_size' )
+    options = parser.parse_args()
+
+    filename = options.output
+
+    params = loads( open( filename ).read() )
+    data_manager_dict = {}
+
+    if options.fasta_dbkey in [ None, '', '?' ]:
+        raise Exception( '"%s" is not a valid dbkey. You must specify a valid dbkey.' % ( options.fasta_dbkey ) )
+
+    sequence_id, sequence_name = get_id_name( params, dbkey=options.fasta_dbkey, fasta_description=options.fasta_description )
+
+    # build the index
+    build_salmon_index( data_manager_dict, options, params, sequence_id, sequence_name )
+
+    # save info to json file
+    open( filename, 'w' ).write( dumps( data_manager_dict ) )
+
+
+if __name__ == "__main__":
+    main()
b
diff -r 6cd60ba8a842 -r c5dea2080109 data_manager/salmon_index_builder.xml
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/data_manager/salmon_index_builder.xml Tue Aug 14 11:19:49 2018 -0400
[
@@ -0,0 +1,38 @@
+<tool id="salmon_index_builder_data_manager" name="Salmon" tool_type="manage_data" version="0.9.1">
+    <description>index builder</description>
+    <requirements>
+        <requirement type="package" version="0.9.1">salmon</requirement>
+    </requirements>
+    <command detect_errors="exit_code"><![CDATA[
+        python '$__tool_directory__/salmon_index_builder.py' --output '${out_file}'
+            --fasta_filename '${all_fasta_source.fields.path}'
+            --fasta_dbkey '${all_fasta_source.fields.dbkey}'
+            --fasta_description '${all_fasta_source.fields.name}'
+            --kmer_size "${kmer_size}"
+            --data_table_name salmon_indexes
+        ]]>
+    </command>
+    <inputs>
+        <param label="Source FASTA Sequence" name="all_fasta_source" type="select">
+            <options from_data_table="all_fasta" />
+        </param>
+        <param name="sequence_name" type="text" value="" label="Name of sequence" />
+        <param name="sequence_id" type="text" value="" label="ID for sequence" />
+        <param name="kmer_size" type="integer" optional='true' value="21" max="32" label="The size of the k-mer on which the index is built"
+                    help="There is a tradeoff here between the distinctiveness of the k-mers and their robustness to errors. The shorter the k-mers, the more robust they will be to errors in the reads, but the longer the k-mers, the more distinct they will be.  We generally recommend using a k-mer size of at least 20. MUST BE AN ODD VALUE "/>
+
+
+
+    </inputs>
+    <outputs>
+        <data name="out_file" format="data_manager_json" />
+    </outputs>
+    <help>
+<![CDATA[
+.. class:: infomark
+
+**Notice:** If you leave name, description, or id blank, it will be generated automatically.
+]]>
+    </help>
+</tool>
+
b
diff -r 6cd60ba8a842 -r c5dea2080109 data_manager_conf.xml
--- a/data_manager_conf.xml Tue Aug 14 11:14:52 2018 -0400
+++ b/data_manager_conf.xml Tue Aug 14 11:19:49 2018 -0400
b
@@ -1,32 +1,17 @@
 <?xml version="1.0"?>
 <data_managers>
-<data_manager tool_file="data_manager/data_manager_fetch_gff.xml" id="data_manager_fetch_gff">
-      <data_table name="all_gff">
+    <data_manager tool_file="data_manager/salmon_index_builder.xml" id="salmon_index_builder" version="0.1">
+        <data_table name="salmon_indexes">
             <output>
                 <column name="value" />
                 <column name="dbkey" />
                 <column name="name" />
-                <column name="path" output_ref="out_file">
-                    <move type="file">
-                        <source>${path}</source>
-                        <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">${dbkey}/gff/${path}</target>
+                <column name="path" output_ref="out_file" >
+                    <move type="directory" relativize_symlinks="True">
+                        <!-- <source>${path}</source>--> <!-- out_file.extra_files_path is used as base by default --> <!-- if no source, eg for type=directory, then refers to base -->
+                        <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">${dbkey}/salmon_index/${value}</target>
                     </move>
-                    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/${dbkey}/gff/${path}</value_translation>
-                    <value_translation type="function">abspath</value_translation>
-                </column>
-            </output>
-        </data_table>
-     <data_table name="representative_gff">
-            <output>
-                <column name="value" />
-                <column name="dbkey" />
-                <column name="name" />
-                <column name="path" output_ref="out_file">
-                    <move type="file">
-                        <source>${path}</source>
-                        <target base="${GALAXY_DATA_MANAGER_DATA_PATH}">${dbkey}/representative_gff/${path}</target>
-                    </move>
-                    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/${dbkey}/representative_gff/${path}</value_translation>
+                    <value_translation>${GALAXY_DATA_MANAGER_DATA_PATH}/${dbkey}/salmon_index/${value}/${path}</value_translation>
                     <value_translation type="function">abspath</value_translation>
                 </column>
             </output>
b
diff -r 6cd60ba8a842 -r c5dea2080109 tool-data/all_fasta.loc.sample
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tool-data/all_fasta.loc.sample Tue Aug 14 11:19:49 2018 -0400
b
@@ -0,0 +1,18 @@
+#This file lists the locations and dbkeys of all the fasta files
+#under the "genome" directory (a directory that contains a directory
+#for each build). The script extract_fasta.py will generate the file
+#all_fasta.loc. This file has the format (white space characters are
+#TAB characters):
+#
+#<unique_build_id> <dbkey> <display_name> <file_path>
+#
+#So, all_fasta.loc could look something like this:
+#
+#apiMel3 apiMel3 Honeybee (Apis mellifera): apiMel3 /path/to/genome/apiMel3/apiMel3.fa
+#hg19canon hg19 Human (Homo sapiens): hg19 Canonical /path/to/genome/hg19/hg19canon.fa
+#hg19full hg19 Human (Homo sapiens): hg19 Full /path/to/genome/hg19/hg19full.fa
+#
+#Your all_fasta.loc file should contain an entry for each individual
+#fasta file. So there will be multiple fasta files for each build,
+#such as with hg19 above.
+#
b
diff -r 6cd60ba8a842 -r c5dea2080109 tool-data/all_gff.loc.sample
--- a/tool-data/all_gff.loc.sample Tue Aug 14 11:14:52 2018 -0400
+++ /dev/null Thu Jan 01 00:00:00 1970 +0000
b
@@ -1,3 +0,0 @@
-#The all_gff.loc file has this format:
-#
-#<unique_build_id> <dbkey> <display_name> <path_to_gff_file>
b
diff -r 6cd60ba8a842 -r c5dea2080109 tool-data/representative_gff.loc.sample
--- a/tool-data/representative_gff.loc.sample Tue Aug 14 11:14:52 2018 -0400
+++ /dev/null Thu Jan 01 00:00:00 1970 +0000
b
@@ -1,3 +0,0 @@
-#The representative_gff.loc file has this format:
-#
-#<unique_build_id> <dbkey> <display_name> <path_to_gff_file>
b
diff -r 6cd60ba8a842 -r c5dea2080109 tool-data/salmon_indexes.loc.sample
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tool-data/salmon_indexes.loc.sample Tue Aug 14 11:19:49 2018 -0400
b
@@ -0,0 +1,28 @@
+# salmon_indexes.loc.sample
+# This is a *.loc.sample file distributed with Galaxy that enables tools
+# to use a directory of indexed data files. This one is for Salmon.
+# See the wiki: http://wiki.galaxyproject.org/Admin/NGS%20Local%20Setup
+# First create these data files and save them in your own data directory structure.
+# Then, create a kallisto_indexes.loc file to use those indexes with tools.
+# Copy this file, save it with the same name (minus the .sample), 
+# follow the format examples, and store the result in this directory.
+# The file should include an one line entry for each index set.
+# The path points to the "basename" for the set, not a specific file.
+# It has four text columns seperated by TABS.
+#
+# <unique_build_id> <dbkey> <display_name> <file_base_path>
+#
+# So, for example, if you had sacCer3 indexes stored in:
+#
+#    /depot/data2/galaxy/sacCer3/salmon_indexes/
+#
+# then the salmon_indexes.loc entry could look like this:
+#
+#sacCer3 sacCer3 S. cerevisiae Apr. 2011 (SacCer_Apr2011/sacCer3) (sacCer3) /depot/data2/galaxy/sacCer3/salmon_indexes
+#
+#More examples:
+#
+#mm10 mm10 Mouse (mm10) /depot/data2/galaxy/salmon_indexes/mm10
+#dm3 dm3 D. melanogaster (dm3) /depot/data2/galaxy/salmon_indexes/dm3
+#
+#
b
diff -r 6cd60ba8a842 -r c5dea2080109 tool_data_table_conf.xml.sample
--- a/tool_data_table_conf.xml.sample Tue Aug 14 11:14:52 2018 -0400
+++ b/tool_data_table_conf.xml.sample Tue Aug 14 11:19:49 2018 -0400
b
@@ -1,5 +1,12 @@
-<?xml version="1.0"?>
 <tables>
- <table name="all_gff" comment_char="#"> <columns>value, dbkey, name, path</columns> <file path="tool-data/all_gff.loc" /> </table>
- <table name="representative_gff" comment_char="#"> <columns>value, dbkey, name, path</columns> <file path="tool-data/representative_gff.loc" /> </table>
+    <!-- Locations of all fasta files under genome directory -->
+    <table name="all_fasta" comment_char="#" allow_duplicate_entries="False">
+        <columns>value, dbkey, name, path</columns>
+        <file path="tool-data/all_fasta.loc" />
+    </table>
+    <!-- Locations of indexes in the kallisto mapper format -->
+    <table name="salmon_indexes" comment_char="#" allow_duplicate_entries="False">
+        <columns>value, dbkey, name, path</columns>
+        <file path="tool-data/salmon_indexes.loc" />
+    </table>
 </tables>