Mercurial > repos > petr-novak > dante

comparison dante.xml @ 10:d0431a839606 draft

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author petr-novak
date Wed, 14 Aug 2019 11:24:15 -0400
parents 6dcecbe81d78
children 3151a72a6671
comparison
equal deleted inserted replaced
9:ed4d9ede9cb4 10:d0431a839606
2 <description> Tool for annotation of transposable elements based on the similarity to conserved protein domains database. </description> 2 <description> Tool for annotation of transposable elements based on the similarity to conserved protein domains database. </description>
3 <requirements> 3 <requirements>
4 <requirement type="package">last</requirement> 4 <requirement type="package">last</requirement>
5 <requirement type="package">numpy</requirement> 5 <requirement type="package">numpy</requirement>
6 <requirement type="package" version="1.0">rexdb</requirement> 6 <requirement type="package" version="1.0">rexdb</requirement>
7 <requirement type="set_environment">REXDB</requirement>
7 </requirements> 8 </requirements>
8 <stdio> 9 <stdio>
9 <regex match="Traceback" source="stderr" level="fatal" description="Unknown error" /> 10 <regex match="Traceback" source="stderr" level="fatal" description="Unknown error" />
10 <regex match="error" source="stderr" level="fatal" description="Unknown error" /> 11 <regex match="error" source="stderr" level="fatal" description="Unknown error" />
11 </stdio> 12 </stdio>
12 <command> 13 <command>
13 python3 ${__tool_directory__}/dante.py --query ${input} --domain_gff ${DomGff} 14 python3 ${__tool_directory__}/dante.py --query ${input} --domain_gff ${DomGff}
14 --protein_database \${REXDB}/${db_type}_pdb 15 --protein_database \${REXDB}/${db_type}_pdb
15 --classification \${REXDB}/${db_type}_class 16 --classification \${REXDB}/${db_type}_class
17 --scoring_matrix ${scoring_matrix}
18 &amp;&amp;
19
20 python3 ${__tool_directory__}/dante_gff_output_filtering.py --dom_gff ${DomGff}
21 --domains_prot_seq domains_filtered.fasta --domains_filtered domains_filtered.gff
22 --output_dir .
23 --selected_dom All --th_identity 0.35
24 --th_similarity 0.45 --th_length 0.9
25 --interruptions 1 --max_len_proportion 1.1
26 --element_type '' &amp;&amp;
27
28 python3 ${__tool_directory__}/fasta2database.py domains_filtered.fasta domains_filtered.db
29 domains_filtered.class &amp;&amp;
30
31 lastdb -p domains_filtered.db domains_filtered.db &amp;&amp;
32
33 python3 ${__tool_directory__}/dante.py --query ${input} --domain_gff ${DomGff2}
34 --protein_database domains_filtered.db
35 --classification domains_filtered.class
36 --scoring_matrix BL80
37
16 </command> 38 </command>
17 <inputs> 39 <inputs>
18 <param format="fasta" type="data" name="input" label="Choose your input sequence" help="Input DNA must be in proper fasta format, multi-fasta containing more sequences is allowed" /> 40 <param format="fasta" type="data" name="input"
41 label="Choose your input sequence" help="Input DNA must be in proper fasta format, multi-fasta containing more sequences is allowed" />
19 42
20 <param name="db_type" type="select" label="Select taxon and protein domain database version (REXdb)" help=""> 43 <param name="db_type" type="select" label="Select taxon and protein domain database version (REXdb)" help="">
21 <options from_file="rexdb_versions.loc"> 44 <options from_file="rexdb_versions.loc">
22 <column name="name" index="0"/> 45 <column name="name" index="0"/>
23 <column name="value" index="1"/> 46 <column name="value" index="1"/>
24 </options> 47 </options>
25 </param> 48 </param>
49
50 <param name="scoring_matrix" type="select" label="Select scoring matrix">
51 <option value="BL80" selected="true" >BLOSUM80</option>
52 <option value="BL62">BLOSUM62</option>
53 <option value="MIQS">MIQS</option>
54 </param>
26 </inputs> 55 </inputs>
27 56
28 <outputs> 57 <outputs>
29 <data format="gff3" name="DomGff" label="Unfiltered GFF3 file of ALL protein domains from dataset ${input.hid}" /> 58 <data format="gff3" name="DomGff" label="protein domains detected in ${input.hid} - 1st pass (unfiltered)" />
59 <data format="gff3" name="DomGff2" label="protein domains detected in ${input.hid} - 2nd pass (unfiltered)" />
30 </outputs> 60 </outputs>
31 <help> 61 <help>
32 62
33 THIS IS A PRIMARY OUTPUT THAT SHOULD UNDERGO FURTHER QUALITY FILTERING TO GET RID OFF POTENTIAL FALSE POSITIVE DOMAINS 63 THIS IS A PRIMARY OUTPUT THAT SHOULD UNDERGO FURTHER QUALITY FILTERING TO GET RID OFF POTENTIAL FALSE POSITIVE DOMAINS
34 64