Skip to content

Commit

Permalink
ADD: misc process files (RNB, CSS, TFA)
Browse files Browse the repository at this point in the history
  • Loading branch information
@pebert
pebert committed Dec 30, 2016
1 parent 6eded56 commit a6efa08
Show file tree
Hide file tree
Showing 4 changed files with 429 additions and 0 deletions.
197 changes: 197 additions & 0 deletions docs/interpretation/CSSv1.xml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,197 @@
<?xml version="1.0"?>
<?xml-stylesheet type="text/css" href="http://deep.mpi-inf.mpg.de/DAC/files/style/deep_process_style.css"?>
<process>
<name>CSS</name>
<version>1</version>
<author>
<name>Peter Ebert</name>
<email>pebert@mpi-inf.mpg.de</email>
</author>
<description>
This process file is in draft status.
The CSS process is designed to generate basic chromatin state segmentation BED files based on histone modifications only.
This process implements a dual strategy to produce the state segmentation: ChromHMM (by Jason Ernst) is used to generate a segmentation
comparable to the reference segmentations provided by the ROADMAP project and thus forms the basis for the automated
state labelling in this process. Additionally, EpiCSeg (by Alessandro Mammana) is used as a more sophisticated method for state
segmentation. Since there is no reference state labelling available for EpiCSeg, the state labels for EpiCSeg are produced via a simple
overlap analysis followed by a majority vote based on the labelled ChromHMM segmentation.
</description>
<inputs>
<filetype>
<identifier>GALvX_Histone</identifier>
<format>BAM</format>
<quantity>collection</quantity>
<comment>To run this process as a default, all six marks need to be available for a sample</comment>
</filetype>
<filetype>
<identifier>GALvX_Input</identifier>
<format>BAM</format>
<quantity>single</quantity>
<comment></comment>
</filetype>
</inputs>
<references>
<filetype>
<identifier>chrom_lengths</identifier>
<format>TXT</format>
<quantity>single</quantity>
<comment>Common 2 column file listing chromosomes (name and length) for assembly</comment>
</filetype>
<filetype>
<identifier>state_labels</identifier>
<format>TXT</format>
<quantity>single</quantity>
<comment>ROADMAP reference state labels (18 states, 6 histone marks)</comment>
</filetype>
<filetype>
<identifier>state_colors</identifier>
<format>TXT</format>
<quantity>single</quantity>
<comment>ROADMAP reference state colors</comment>
</filetype>
<filetype>
<identifier>blacklist_regions</identifier>
<format>BED</format>
<quantity>single</quantity>
<comment>Common blacklist regions, same as for CHP or DHS</comment>
</filetype>
<filetype>
<identifier>var_ref_files</identifier>
<format>BED</format>
<quantity>collection</quantity>
<comment>It is still undecided if a fix set of annotation files should be used for the
reporting feature of the tools (state overlap); due to the automated state labelling procedure,
this is not necessary</comment>
</filetype>
</references>
<outputs>
<filetype>
<identifier>state_segmentation</identifier>
<format>BED</format>
<quantity>collection</quantity>
<comment>segmentations procduced by ChromHMM and EpiCSeg and post-processed to be properly labelled</comment>
</filetype>
<filetype>
<identifier>model</identifier>
<format>TXT</format>
<quantity>collection</quantity>
<comment>EpiCSeg model</comment>
</filetype>
<filetype>
<identifier>reports</identifier>
<format>tar.gz</format>
<quantity>collection</quantity>
<comment>combined set of all other output files that are part of the tool's report</comment>
</filetype>
</outputs>
<software>
<tool>
<name>samtools</name>
<version>1.2</version>
<command_line><![CDATA[ samtools view -b -F 1024 {GALvX_*} > tmp_nodup.bam ]]></command_line>
<loop>GALvX_Histone, GALvX_Input</loop>
<comment>Remove duplicates</comment>
</tool>
<tool>
<name>samtools</name>
<version>1.2</version>
<command_line><![CDATA[ samtools sort -n -T tmpsrt -O bam tmp_nodup.bam ]]></command_line>
<loop>GALvX_Histone, GALvX_Input</loop>
<comment>Sort prior to filtering blacklist, output is piped to next command</comment>
</tool>
<tool>
<name>bedtools</name>
<version>2.20.1</version>
<command_line><![CDATA[ pairToBed -ubam -type neither -abam - -b {blacklist_regions} ]]></command_line>
<loop></loop>
<comment>Filter blacklist regions, output is piped to next command</comment>
</tool>
<tool>
<name>samtools</name>
<version>1.2</version>
<command_line><![CDATA[ samtools sort -o tmp_nodup_blfilt.bam -T tmpsrt2 -O bam - && samtools index tmp_nodup_blfilt.bam ]]></command_line>
<loop></loop>
<comment>Create filtered and indexed BAM file</comment>
</tool>
<tool>
<name>bedtools</name>
<version>2.20.1</version>
<command_line><![CDATA[ bedtools bamtobed -i tmp_nodup_blfilt.bam | egrep "^(chr)?[0-9XY]+\s" > tmp_nodup_blfilt.bed ]]></command_line>
<loop></loop>
<comment>Make simple BED file as input for ChromHMM</comment>
</tool>
<tool>
<name>java, ChromHMM.jar</name>
<version>1.7.0_65, 1.10</version>
<command_line><![CDATA[ ChromHMM.jar BinarizeBed {chrom_lengths} wrk_tmp_dir cellmarktable wrk_tmp_dir ]]></command_line>
<loop></loop>
<comment>The cellmarktable info file is built on-the-fly by the pipeline and discarded after the run</comment>
</tool>
<tool>
<name>java, ChromHMM.jar</name>
<version>1.7.0_65, 1.10</version>
<command_line><![CDATA[ ChromHMM.jar MakeSegmentation -printposterior -b {bin_size} {remc_model} wrk_tmp_dir wrk_out_dir ]]></command_line>
<loop></loop>
<comment>The cellmarktable info file is built on-the-fly by the pipeline and discarded after the run</comment>
</tool>
<tool>
<name>R, EpiCSeg</name>
<version>3.2.0, 2016-04-04</version>
<command_line><![CDATA[ Rscript epicseg.R getcounts --nthreads {ecs_parallel} --pairedend TRUE --regions tmp_regions --target {DEEPID.counts} -m tmp_nodup_blfilt.bam ]]></command_line>
<loop>no looping</loop>
<comment>The command is assembled via repeating the -m parameter (-m Mark:Filepath) for the marks/Input</comment>
</tool>
<tool>
<name>R, EpiCSeg</name>
<version>3.2.0, 2016-04-04</version>
<command_line><![CDATA[ Rscript epicseg.R normalizecounts --nthreads {ecs_parallel} -c DEEPID.counts ]]></command_line>
<loop>no looping</loop>
<comment>This normalization step is mandatory if a joined segmentation is done for several samples (e.g. for a whole sub-project);
in the default case of a single sample, this has no relevant effect</comment>
</tool>
<tool>
<name>R, EpiCSeg</name>
<version>3.2.0, 2016-04-04</version>
<command_line><![CDATA[ Rscript epicseg.R segment --nthreads {ecs_parallel} --nstates 18 --outdir wrk_out_dir --regions tmp_regions -c DEEPID.counts.norm ]]></command_line>
<loop>no looping</loop>
<comment></comment>
</tool>
<tool>
<name>bedtools</name>
<version>2.20.1</version>
<command_line><![CDATA[ bedtools intersect -wo -a {ecs_segmentation} -b {cmm_segmentation}} > ecs_cmm_isect.tmp ]]></command_line>
<loop>EpiCSeg and ChromHMM segmentations</loop>
<comment></comment>
</tool>
<tool>
<name>Python3</name>
<version>3.2.3</version>
<command_line><![CDATA[ process_css.py --task corres -i ecs_cmm_isect.tmp -o ecs_cmm_state_corres.tsv]]></command_line>
<loop>no looping</loop>
<comment>Assign labels for the EpiCSeg segmentation based on overlap with ChromHMM segmentation. This command outputs a label mapping file</comment>
</tool>
<tool>
<name>Python3</name>
<version>3.2.3</version>
<command_line><![CDATA[ process_css.py --task track -i ecs_segment.tmp -o {state_segmentation} --state-corres ecs_cmm_state_corres.tsv --track-name "SAMPLEID (ECS)" --track-desc "SAMPLEID CSSvX EpiCSeg"]]></command_line>
<loop>no looping</loop>
<comment>Assign labels for the EpiCSeg segmentation based on overlap with ChromHMM segmentation. Produces a browsertrack BED file to be loaded into, e.g., IGV</comment>
</tool>
<tool>
<name>Python3</name>
<version>3.2.3</version>
<command_line><![CDATA[ process_css.py --task track -i cmm_segment.tmp -o {state_segmentation} --color-map {state_colors} --label-map {state_labels} --track-name "SAMPLEID (CMM)" --track-desc "SAMPLEID CSSvX ChromHMM"]]></command_line>
<loop>no looping</loop>
<comment>Assign labels for the EpiCSeg segmentation based on overlap with ChromHMM segmentation. Produces a browsertrack BED file to be loaded into, e.g., IGV</comment>
</tool>
<tool>
<name>Unspec</name>
<version>0.0</version>
<command_line><![CDATA[ package model outputs ]]></command_line>
<loop>no looping</loop>
<comment>To do command for packaging the various output files into 2 gzip files</comment>
</tool>
</software>
</process>


59 changes: 59 additions & 0 deletions docs/interpretation/RNBv0.xml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,59 @@
<?xml version="1.0"?>
<?xml-stylesheet type="text/css" href="http://deep.mpi-inf.mpg.de/DAC/files/style/deep_process_style.css"?>
<process>
<name>RNB</name>
<version>0</version>
<author>
<name>Fabian Mueller</name>
<email>fmueller@mpi-inf.mpg.de</email>
</author>
<description>
This process desribes an integrative analysis of bisulfite methylation data. The input comprises methylation calls in bed format, a comma-separated sample annotation file and options setting in XML format. The result is a folder structure of multiple HTML reports describing data loading, quality control, site and sample filtering, identification of batch effects, methylation profiling on the basis of indiviual CpGs as well as genomic regions, differential methylation analysis and data export.
These reports can be viewed locally or displayed via the internet.
</description>
<inputs>
<filetype>
<identifier>MCSv0.bed</identifier>
<format></format>
<quantity>collection</quantity>
<comment>methylation calls in bed format</comment>
</filetype>
<filetype>
<identifier>sampleAnnotation.csv</identifier>
<format></format>
<quantity>single</quantity>
<comment>Sample annotation table</comment>
</filetype>
<filetype>
<identifier>analysisOptions.xml</identifier>
<format></format>
<quantity>single</quantity>
<comment>Options settings for RnBeads. These options are also listed in the corresponding analysis metadata. </comment>
</filetype>
</inputs>
<references>
<filetype>
<identifier>methylationCall.bed</identifier>
<format></format>
<quantity>collection</quantity>
<comment>Methylation calls from other samples, not obtained by DEEP in bed format.</comment>
</filetype>
</references>
<outputs>
<filetype>
<identifier>RnBv0.tar.gz</identifier>
<format></format>
<quantity>single</quantity>
<comment>RnBeads report directory containing HTML reports and associated plots, tables and files.</comment>
</filetype>
</outputs>
<software>
<tool>
<name>RnBeads</name>
<version>0.99.11</version>
<command_line><![CDATA[ R --vanilla --sampleAnnotation=sampleAnnotation.csv --analysisOptions=analysisOptions.xml < runRnBeads.R ]]> </command_line>
<loop></loop>
<comment>Integrated analysis of multiple samples using RnBeads. Pipeline options are transferred to RnBeads via an XML file, but should also be specified in the anlysis metadata.</comment>
</tool>
</software>
</process>
47 changes: 47 additions & 0 deletions docs/interpretation/RNBv1.xml
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,47 @@
<?xml version="1.0"?>
<?xml-stylesheet type="text/css" href="http://deep.mpi-inf.mpg.de/DAC/files/style/deep_process_style.css"?>
<process>
<name>RNB</name>
<version>1</version>
<author>
<name>Fabian Mueller</name>
<email>fmueller@mpi-inf.mpg.de</email>
</author>
<description>
This process desribes an integrative analysis of bisulfite methylation data. The analysis is run from a configuration JSON file. The input comprises methylation calls in bed format, a comma-separated sample annotation file and options setting in XML format which are specified in the analysis metadata. The result is a folder structure of multiple HTML reports describing data loading, quality control, site and sample filtering, identification of batch effects, methylation profiling on the basis of indiviual CpGs as well as genomic regions, differential methylation analysis and data export.
These reports can be viewed locally or displayed via the internet.
</description>
<inputs>
<filetype>
<identifier>ANALYSIS_CONFIG.JSON</identifier>
<format></format>
<quantity>single</quantity>
<comment>Options settings for RnBeads. These options are also listed in the corresponding analysis metadata.</comment>
</filetype>
</inputs>
<references>
<filetype>
<identifier>species_annotation</identifier>
<format>R package</format>
<quantity>single</quantity>
<comment>Assembly specific annotation package containing standard information like CpG islands, promoter regions etc, see http://rnbeads.mpi-inf.mpg.de/installation.php</comment>
</filetype>
</references>
<outputs>
<filetype>
<identifier>RNBEADS_REPORT</identifier>
<format></format>
<quantity>single</quantity>
<comment>RnBeads report directory containing HTML reports and associated plots, tables and files.</comment>
</filetype>
</outputs>
<software>
<tool>
<name>RnBeads</name>
<version>>=1.1.3</version>
<command_line><![CDATA[ Rscript run_rnbeads.R {ANALYSIS_CONFIG.JSON}]]></command_line>
<loop></loop>
<comment>Integrated analysis of multiple samples using RnBeads. Pipeline options are transferred to RnBeads via a JSON file, but should also be specified in the anlysis metadata.</comment>
</tool>
</software>
</process>
Loading

0 comments on commit a6efa08

Please sign in to comment.