# HG changeset patch
# User morinlab
# Date 1476210085 14400
# Node ID 3f5545f5f0c8a40552ed49708608cc25beda2fd7
planemo upload for repository https://github.com/morinlab/tools-morinlab/tree/master/tools/mutationseq commit 4ef2d91b7c1686a2696b92fe538d4aec51d05e40-dirty
diff -r 000000000000 -r 3f5545f5f0c8 citations.xml
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/citations.xml Tue Oct 11 14:21:25 2016 -0400
@@ -0,0 +1,153 @@
+
+
+
+ @unpublished{
+ albuquerque2016galaxy,
+ author = "Marco Albuquerque and Bruno Grande and Elie Ritch and Martin Krzywinski and Prasath Pararajalingam and Selin Jessa and Paul Boutros and Sohrab Shah and Ryan Morin",
+ title = "A Suite of Galaxy Tools for Cancer Mutational Analysis",
+ note = "Unpublished Manuscript",
+ year = "2016"
+ }
+
+
+
+
+ @article{
+ Lai20062016,
+ title = {VarDict: a novel and versatile variant caller for next-generation sequencing in cancer research},
+ author = {Lai, Zhongwu and Markovets, Aleksandra and Ahdesmaki, Miika and Chapman, Brad and Hofmann, Oliver and McEwen, Robert and Johnson, Justin and Dougherty, Brian and Barrett, J. Carl and Dry, Jonathan R.},
+ journal = {Nucleic Acids Research}
+ volume = {44},
+ number = {11},
+ pages = {e108},
+ year = {2016},
+ doi = {10.1093/nar/gkw227}
+ }
+
+
+
+
+ @article{
+ Larson01022012,
+ author = {Larson, David E. and Harris, Christopher C. and Chen, Ken and Koboldt, Daniel C. and Abbott, Travis E. and Dooling, David J. and Ley, Timothy J. and Mardis, Elaine R. and Wilson, Richard K. and Ding, Li},
+ title = {SomaticSniper: identification of somatic point mutations in whole genome sequencing data},
+ volume = {28},
+ number = {3},
+ pages = {311-317},
+ year = {2012},
+ doi = {10.1093/bioinformatics/btr665},
+ journal = {Bioinformatics}
+ }
+
+
+
+
+ @ARTICLE{Goecks2010-ra,
+ title = "Galaxy: a comprehensive approach for supporting accessible,
+ reproducible, and transparent computational research in the life
+ sciences",
+ author = "Goecks, Jeremy and Nekrutenko, Anton and Taylor, James and
+ {Galaxy Team}",
+ journal = "Genome Biol.",
+ volume = 11,
+ number = 8,
+ pages = "R86",
+ month = "25~",
+ year = 2010
+ }
+
+
+
+
+
+ @ARTICLE{Rausch2012-yi,
+ title = "{DELLY}: structural variant discovery by integrated paired-end
+ and split-read analysis",
+ author = "Rausch, Tobias and Zichner, Thomas and Schlattl, Andreas and
+ St{\"{u}}tz, Adrian M and Benes, Vladimir and Korbel, Jan O",
+ journal = "Bioinformatics",
+ volume = 28,
+ number = 18,
+ pages = "i333--i339",
+ month = "15~",
+ year = 2012
+}
+
+
+
+
+@ARTICLE{Ding2012-jq,
+ title = "Feature-based classifiers for somatic mutation detection in
+ tumour-normal paired sequencing data",
+ author = "Ding, Jiarui and Bashashati, Ali and Roth, Andrew and Oloumi,
+ Arusha and Tse, Kane and Zeng, Thomas and Haffari, Gholamreza and
+ Hirst, Martin and Marra, Marco A and Condon, Anne and Aparicio,
+ Samuel and Shah, Sohrab P",
+ journal = "Bioinformatics",
+ volume = 28,
+ number = 2,
+ pages = "167--175",
+ month = "15~" # jan,
+ year = 2012
+}
+
+
+
+
+
+ @ARTICLE{Saunders2012-nh,
+ title = "Strelka: accurate somatic small-variant calling from sequenced
+ tumor-normal sample pairs",
+ author = "Saunders, Christopher T and Wong, Wendy S W and Swamy, Sajani and
+ Becq, Jennifer and Murray, Lisa J and Cheetham, R Keira",
+ journal = "Bioinformatics",
+ volume = 28,
+ number = 14,
+ pages = "1811--1817",
+ month = "15~" # jul,
+ year = 2012
+ }
+
+
+
+
+ @article{
+ Radenbaugh2014-tj,
+ title={RADIA: RNA and DNA integrated analysis for somatic
+ mutation detection},
+ author={Radenbaugh, Amie J and Ma, Singer and Ewing, Adam and Stuart,
+ Joshua M and Collisson, Eric A and Zhu, Jingchun and Haussler,
+ David},
+ journal={PLoS One},
+ volume={9},
+ number={11},
+ pages={e111516},
+ year={2014},
+ publisher={PLoS}
+ }
+
+
+
+
+ @ARTICLE{Ha2014-pu,
+ title = "{TITAN}: inference of copy number architectures in clonal cell
+ populations from tumor whole-genome sequence data",
+ author = "Ha, Gavin and Roth, Andrew and Khattra, Jaswinder and Ho,
+ Julie and Yap, Damian and Prentice, Leah M and Melnyk,
+ Nataliya and McPherson, Andrew and Bashashati, Ali and Laks,
+ Emma and Biele, Justina and Ding, Jiarui and Le, Alan and
+ Rosner, Jamie and Shumansky, Karey and Marra, Marco A and
+ Gilks, C Blake and Huntsman, David G and McAlpine, Jessica N
+ and Aparicio, Samuel and Shah, Sohrab P",
+ journal = "Genome Res.",
+ publisher = "Cold Spring Harbor Lab",
+ volume = 24,
+ number = 11,
+ pages = "1881--1893",
+ month = "1~" # nov,
+ year = 2014,
+ keywords = "computational method"
+}
+
+
+
diff -r 000000000000 -r 3f5545f5f0c8 mutationseq.xml
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/mutationseq.xml Tue Oct 11 14:21:25 2016 -0400
@@ -0,0 +1,174 @@
+
+
+ a feature based classifier for somatic mutation detection
+
+
+ mutationseq
+ mutationseq_python_environment
+
+
+ citations.xml
+
+
+
+
+ #if $model_source.model_source_selector == "standard" or $model_source.model_source_selector == "solid" or $model_source.model_source_selector == "deepseq" or $model_source.model_source_selector == "titan_mode"
+ ln -s $model_source.normal normal.bam;
+ ln -s $model_source.normal.metadata.bam_index normal.bam.bai;
+ ln -s $model_source.tumour tumour.bam;
+ ln -s $model_source.tumour.metadata.bam_index tumour.bam.bai;
+
+ #else
+ ln -s $model_source.input input.bam;
+ ln -s $model_source.input.metadata.bam_index input.bam.bai;
+
+ #end if
+
+
+ #if $reference_source.reference_source_selector == "history":
+ ln -s $reference_source.ref_file ref.fa;
+ samtools faidx ref.fa;
+ #end if
+
+
+ #if $interval:
+ for i in \$(cat $interval); do
+ #end if
+
+ #if $model_source.model_source_selector == "titan_mode":
+ \$MUTATIONSEQ_PYTHON_ENVIRONMENT_ROOT_DIR/bin/python2.7 \$MUTATIONSEQ_ROOT_DIR/preprocess.py
+ #else:
+ \$MUTATIONSEQ_PYTHON_ENVIRONMENT_ROOT_DIR/bin/python2.7 \$MUTATIONSEQ_ROOT_DIR/classify.py
+ #end if
+
+ #if $reference_source.reference_source_selector == "history":
+ reference:ref.fa
+ #else:
+ reference:${reference_source.ref_file.fields.path}
+ #end if
+
+ #if $model_source.model_source_selector == "titan_mode":
+ normal:normal.bam
+ tumour:tumour.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_single_v4.0.2.npz
+ --single
+
+ #elif $model_source.model_source_selector == "standard"
+ normal:normal.bam
+ tumour:tumour.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_v4.1.2.npz
+
+ #elif $model_source.model_source_selector == "standard_single"
+ normal:input.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_single_v4.0.2.npz
+ --single
+
+ #elif $model_source.model_source_selector == "solid"
+ normal:normal.bam
+ tumour:tumour.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_solid_v4.1.2.npz
+
+ #elif $model_source.model_source_selector == "deepseq"
+ normal:normal.bam
+ tumour:tumour.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_deep_v0.2.npz
+ --deep
+
+ #elif $model_source.model_source_selector == "deepseq_single"
+ normal:input.bam
+ model:\$MUTATIONSEQ_ROOT_DIR/model_deep_single_v0.2.npz
+ --single
+ --deep
+
+ #end if
+
+ #if $advancedsettings.filter.value == "all"
+ --all
+ #elif $advancedsettings.filter.value == "no_filter"
+ --no_filter
+ #end if
+
+ #if $interval
+ -i \$i
+ #end if
+
+ #if $advancedsettings.positions:
+ -f $advancedsettings.positions
+ #end if
+
+ --threshold $advancedsettings.filter_prob
+
+ | grep -v ".*no somatic mutation calls.*" >> $output
+
+ #if $interval
+ ; done
+ #end if
+ ;
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff -r 000000000000 -r 3f5545f5f0c8 tool-data/all_fasta.loc.sample
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tool-data/all_fasta.loc.sample Tue Oct 11 14:21:25 2016 -0400
@@ -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):
+#
+#
+#
+#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.
+#
diff -r 000000000000 -r 3f5545f5f0c8 tool_data_table_conf.xml.sample
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/tool_data_table_conf.xml.sample Tue Oct 11 14:21:25 2016 -0400
@@ -0,0 +1,6 @@
+
+