Mercurial > repos > fubar > blasttools_search_test

comparison blasttoolssearch/blasttoolssearch.xml @ 0:ee581a90a85e draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded initial version
author fubar
date Wed, 19 Jul 2023 04:34:01 +0000
parents
children 5687b8f1ad69
comparison
equal deleted inserted replaced
-1:000000000000 0:ee581a90a85e
1 <tool name="blasttoolssearch" id="blasttoolssearch" version="3.0">
2 <!--Source in git at: https://github.com/fubar2/galaxy-->
3 <!--Created by toolfactory@galaxy.org at 19/07/2023 12:39:19 using the Galaxy Tool Factory.-->
4 <description>Runs a legacy Java jar called blasttools from https://github.com/schmidda/blast-tools/tree/master</description>
5 <requirements>
6 <requirement type="package">csvtk</requirement>
7 <requirement type="package">openjdk</requirement>
8 </requirements>
9 <stdio>
10 <exit_code range="1:" level="fatal"/>
11 </stdio>
12 <version_command><![CDATA[echo "3.0"]]></version_command>
13 <command><![CDATA[bash
14 $runme
15 $blastn_search_outputs
16 $__tool_directory__/BlastTools.jar
17 $summary_viruses_viroids
18 ]]></command>
19 <configfiles>
20 <configfile name="runme"><![CDATA[#raw
21
22
23 ## eResearch Office, QUT
24 ## Created: 31 March 2021
25 ## Last modified: 28 September 2022
26 ## Script: Processes Galaxy Australia generated blastN outputs to summarise and report hits to REGULATED and ENDEMIC viruses/viroids.
27 ## Usage: ./run_VirReport_Summary.sh
28 ## changed to accept a single input file name passed as $1
29 ## Ross Lazarus for a ToolFactory wrapper for Robert Barrero
30 ## July 18 2023
31
32 dataPath=${PWD}
33
34 # Requirement: One or more GA-VSD .tabular outputs need to be in the folder where the command above (Usage)is executed.
35 # The script will Look for all files with the suffix *.tabular
36
37 #Processing tabular files
38 file=$1
39
40 var=$(basename $file)
41
42 #STEP0: fetch Top 1 Hits
43 cat $file | awk '{print $1}' | sort | uniq > ${var}.top1.ids
44 for i in `cat ${var}.top1.ids`
45 do
46 echo "fetching top hits..." $i;
47 grep $i $file | head -1 >> ${var}.top1Hits.txt;
48 done
49
50 #STEP1: modify the columns of Galaxy Australia (GA) blast output to the expected format by the BlastTools.jar tool
51 ###### namely: qseqid sgi sacc length pident mismatch gapopen qstart qend qlen sstart send slen sstrand evalue bitscore qcovhsp stitle staxids qseq sseq sseqid qcovs qframe sframe
52 cat ${var}.top1Hits.txt |csvtk cut -H -t -f 1,19,20,4,3,5,6,7,8,17,9,10,18,22,11,12,24,21,25,15,16,2,23,13,14 | sed 's/ /_/g' > ${var}.txt
53
54 #STEP2: summarise the GA blastN files
55 java -jar $2 -t blastn ${var}.txt
56 #filter virus/viroid/endo
57 cat summary_${var}.txt | grep "virus\|viroid\|endo" > summary_${var}_filtered.txt
58
59 #STEP3: fetch unique names from Blast summary reports
60 cat summary_${var}_filtered.txt | awk '{print $7}' | awk -F "|" '{print $2}'| sort | uniq | sed 's/Species://' > ${var}_uniq.ids
61
62 #STEP4: retrieve the best hit for each virus/viroid
63 echo "processing top hits ..."
64 touch ${var}_filtered.txt
65 for id in `cat ${var}_uniq.ids`
66 do
67 #print on the screen the name of the virus/viroids to search
68 #echo "fetching species matches ..." $id
69
70 #fetch the virus name on the summary_blastn file by selecting the longest alignment (column 3) and highest genome coverage (column 5)
71 grep $id summary_${var}.txt | sort -k3,3nr -k5,5nr | head -1 >> ${var}_filtered.txt
72 done
73
74 #print the header of the inital summary_blastn file
75 cat summary_${var}.txt | head -1 > header
76 #report 1
77 cat header ${var}_filtered.txt > $3
78
79 #end raw]]></configfile>
80 </configfiles>
81 <inputs>
82 <param name="blastn_search_outputs" type="data" optional="false" label="blastn_search_outputs" help="" format="tabular" multiple="false"/>
83 </inputs>
84 <outputs>
85 <data name="summary_viruses_viroids" format="txt" label="summary_viruses_viroids" hidden="false"/>
86 </outputs>
87 <tests>
88 <test>
89 <output name="summary_viruses_viroids" value="summary_viruses_viroids_sample" compare="diff" lines_diff="0"/>
90 <param name="blastn_search_outputs" value="blastn_search_outputs_sample"/>
91 </test>
92 </tests>
93 <help><![CDATA[
94
95 **What it Does**
96
97 Wraps https://github.com/schmidda/blast-tools/tree/master as a Galaxy tool as a demonstration for Roberto Barrero
98
99
100
101 ------
102
103
104 Script::
105
106 ## eResearch Office, QUT
107 ## Created: 31 March 2021
108 ## Last modified: 28 September 2022
109 ## Script: Processes Galaxy Australia generated blastN outputs to summarise and report hits to REGULATED and ENDEMIC viruses/viroids.
110 ## Usage: ./run_VirReport_Summary.sh
111 ## changed to accept a single input file name passed as $1
112 ## Ross Lazarus for a ToolFactory wrapper for Robert Barrero
113 ## July 18 2023
114 dataPath=${PWD}
115 # Requirement: One or more GA-VSD .tabular outputs need to be in the folder where the command above (Usage)is executed.
116 # The script will Look for all files with the suffix *.tabular
117 #Processing tabular files
118 file=$1
119 var=$(basename $file)
120 #STEP0: fetch Top 1 Hits
121 cat $file | awk '{print $1}' | sort | uniq > ${var}.top1.ids
122 for i in `cat ${var}.top1.ids`
123 do
124 echo "fetching top hits..." $i;
125 grep $i $file | head -1 >> ${var}.top1Hits.txt;
126 done
127 #STEP1: modify the columns of Galaxy Australia (GA) blast output to the expected format by the BlastTools.jar tool
128 ###### namely: qseqid sgi sacc length pident mismatch gapopen qstart qend qlen sstart send slen sstrand evalue bitscore qcovhsp stitle staxids qseq sseq sseqid qcovs qframe sframe
129 cat ${var}.top1Hits.txt |csvtk cut -H -t -f 1,19,20,4,3,5,6,7,8,17,9,10,18,22,11,12,24,21,25,15,16,2,23,13,14 | sed 's/ /_/g' > ${var}.txt
130 #STEP2: summarise the GA blastN files
131 #java -jar /mnt/c/Users/lelwala/HTS/BlastTools.jar -t blastn ${var}.txt
132 java -jar $3 -t blastn ${var}.txt
133 #filter virus/viroid/endo
134 cat summary_${var}.txt | grep "virus\|viroid\|endo" > summary_${var}_filtered.txt
135 #STEP3: fetch unique names from Blast summary reports
136 cat summary_${var}_filtered.txt | awk '{print $7}' | awk -F "|" '{print $2}'| sort | uniq | sed 's/Species://' > ${var}_uniq.ids
137 #STEP4: retrieve the best hit for each virus/viroid
138 echo "processing top hits ..."
139 touch ${var}_filtered.txt
140 for id in `cat ${var}_uniq.ids`
141 do
142 #print on the screen the name of the virus/viroids to search
143 #echo "fetching species matches ..." $id
144 #fetch the virus name on the summary_blastn file by selecting the longest alignment (column 3) and highest genome coverage (column 5)
145 grep $id summary_${var}.txt | sort -k3,3nr -k5,5nr | head -1 >> ${var}_filtered.txt
146 done
147 #print the header of the inital summary_blastn file
148 cat summary_${var}.txt | head -1 > header
149 #report 1
150 cat header ${var}_filtered.txt > $2
151 #removing intermediate files
152 rm summary_${file}.txt ${file}.txt ${file}.top1.ids ${file}_uniq.ids summary_${file}_filtered.txt header* ${var}_filtered.txt *top1Hits.txt
153
154 ]]></help>
155 <citations>
156 <citation type="doi">10.1093/bioinformatics/bts573</citation>
157 </citations>
158 </tool>
159