+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+ /tools-devteam
+
+
+
+
+
+ -
+
+
- + + + +
- + + + + + +
- + + +
+
+ /tools-devteam
+
+
+ 
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
++ + +
+The reusable macro file has: + + - Defines widgets for common read group parameters. + - Widgets can be used at top-level (e.g. ``picard_AddOrReplaceReadGroups`` where handling is required) of inputs or within a conditional (e.g. in the mappers). + - The conditional (used by mappers) has fours modes: + - Set according the specification (only ID required - has all parameters). Current BWA approach. + - Set Picard-style (presumably useful for broad pipelines?) (``ID``, ``SM``, ``LB``, ``PU``, ``PL`` are all required and there are no params for ``KS``, ``PG``, or ``FO``). In terms of what is shown and what is required - this mirrors ``picard_AddOrReplaceReadGroups``. + - Just assign the ``ID`` and do so automatically (based either on collection or standalone dataset information). No extra form elements are shown to the user. (Should this be the new default?) + - Do not set read groups. + - Utilities for conditionally including parameters and formatting them (eliminates many explicit conditionals in the mappers for instance). + - An example "mapper" test tool which demonstrates optionally including read groups. + - An example read group assign test tool which demonstrates unconditionally specifying macros. + +In addition to unifying everything, the macro file defines enhanced ``ID``, ``SM``, and ``LB`` parameter inputs that are wrapped in a conditional. That conditional allows manual input but also allows the user to just say Galaxy should use collection information or dataset name (if input is not in a collection) to assign these parameters. This should simplify things and make simple workflows more robust, but is absolutely essential for collection-based workflows and newer tool form options that allow running a tool in parallel over many inputs. + +This may look at first glance like it is really complicated - but if you focus just on the mappers for instance and treat the macro file like a library - the client interface is actually a good deal easier than doing this stuff by hand for each tool. + +Updates from original version: + + - Tests pass for ``picard_AddOrReplaceReadGroups`` as well as the BWA and Bowtie tools. + - Remove the changes from tophat - it will require more work (it is less free form and won't accept certain valid read groups arguments). + - Deleted empty lines I had previously introduced and idented the Cheetah code (hadn't realized I could do that). + +Smaller Notes: + +The new approach for BWA tools excludes empty read groups instead of adding them as empty - hopefully this is fine: + +``` +-@RG ID:rg1 SM: PL:CAPILLARY +-@PG ID:bwa PN:bwa VN:0.7.12-r1039 CL:bwa sampe -r @RG ID:rg1 SM: PL:CAPILLARY localref.fa first.sai second.sai /tmp/tmpn_I5mQ/files/000/dataset_9.dat /tmp/tmpn_I5mQ/files/000/dataset_10.dat ++@RG ID:rg1 PL:CAPILLARY ++@PG ID:bwa PN:bwa VN:0.7.12-r1039 CL:bwa sampe -r @RG ID:rg1 PL:CAPILLARY localref.fa first.sai second.sai /tmp/tmpmN3j1j/files/000/dataset_9.dat /tmp/tmpmN3j1j/files/000/dataset_10.dat +```
+
+ latest commit d0e3412c58
+
+
+
+ 
+
+ authored
+
+
+
+
+ authored
+
+
+ Permalink
+
+
+
+
+
+
+
+
+
+
+ | + | .. | ++ | + |
| + | Failed to load latest commit information. | +||
|
+
+
+ |
+ + bowtie2-fq1.fq + | + ++ + | +|
|
+
+
+ |
+ + bowtie2-fq2.fq + | + ++ + | +|
|
+
+
+ |
+ + bowtie2-ref.fasta + | + ++ + | +|
|
+
+
+ |
+ + bowtie2-test1.bam + | + ++ + | +|
|
+
+
+ |
+ + bowtie2-test2.bam + | + ++ + | +|
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff -r 18b04f0fb939 -r 8accf8976cf7 url2bowtie/tool_data_table_conf.xml.sample
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/url2bowtie/tool_data_table_conf.xml.sample Mon Aug 03 05:36:36 2015 -0400
@@ -0,0 +1,8 @@
+
+