Brad et al, I would like second the issue you raise so succinctly. The failure to automatically track the original sample name throughout the analysis (that and array selection of paired end reads) is one of the biggest barriers people face for doing work on many samples in galaxy. It just gets very confusing unless you spend a lot of time workarounds (creating workflows to rename things, editing datasets individually, etc) – especially for non-programmer users, for whom workflows with variables and API calls are beyond the pale. Regards, Curtis From: galaxy-dev-bounces@lists.bx.psu.edu [mailto:galaxy-dev-bounces@lists.bx.psu.edu] On Behalf Of Langhorst, Brad Sent: Thursday, November 13, 2014 3:23 PM To: Dannon Baker Cc: galaxy-dev@lists.bx.psu.edu Subject: Re: [galaxy-dev] rename output dataset in workflow - input dataset variable I think adding the available inputs is a good improvement - I know it will save me some mistakes. I think I might have been unclear in my comment ("single least intuitive part")... I'm not referring to how input name chaining works - I'm talking about the default naming of datasets based on the analysis step and the immediate inputs. When interpreting results, we care about two things: 1) which analysis 2) which sample is being described by that analysis. The chain of analysis is important, but its easily determined by looking at the info tab, or generating a workflow from the history to the order of events. The current naming system requires that we walk the entire chain of analysis back to the original dataset, or laboriously build every workflow to manually carry the sample information along in the input and output naming. If the default display convention showed the original dataset dataset name, much of that toting of input name could be avoided. Maybe I'll find some time to implement this someday... Brad ________________________________ From: Dannon Baker <dannon.baker@gmail.com<mailto:dannon.baker@gmail.com>> Sent: Thursday, November 13, 2014 10:05 AM To: Langhorst, Brad Cc: Jan Hapala; galaxy-dev@lists.bx.psu.edu<mailto:galaxy-dev@lists.bx.psu.edu> Subject: Re: [galaxy-dev] rename output dataset in workflow - input dataset variable Hi all, For these parameters, 'input' refers to the exact tool input name. Recent versions of Galaxy will now tell you what the options are to use, like the following I see for bowtie2: [Inline image 1] I definitely agree that it's not intuitive, and I'm trying to think of ways to make this better -- suggestions definitely welcome. -Dannon On Wed, Nov 12, 2014 at 6:54 AM, Langhorst, Brad <Langhorst@neb.com<mailto:Langhorst@neb.com>> wrote: Hi Jan: Are you using input dataset tool? I'm not sure it's required - but I habitually use it. I'm attaching an image of a working renaming setup. I hope it helps. brad [cid:image002.png@01D00405.AA7E35D0] ________________________________ From: Jan Hapala <jan@hapala.cz<mailto:jan@hapala.cz>> Sent: Wednesday, November 12, 2014 9:29 AM To: Langhorst, Brad Subject: Re: [galaxy-dev] rename output dataset in workflow - input dataset variable Thanks, Brad. This is informative, but I'm still stuck. I have an input dataset which passes data to FASTQ Summary Statistics. (I have tried other tools, as well.) The output file name is: stats on: #{input1} The name of the Input dataset is: input1 I get this result: stats on: Jan 2014-11-12 14:06 GMT+01:00 Langhorst, Brad <Langhorst@neb.com<mailto:Langhorst@neb.com>>: For what it's worth (not much ;), I think the dataset naming is the single least intuitive part of the galaxy system for biologists. I think there's a proposal to "fix" this problem - and mainain the connection between input data set and derived data, but It doesn't seem to have gone anywhere yet. Anyway, you have to check the name of the field for each tool - it's sometimes "input" and sometimes something else. You want to use the #{} for naming. The ${} stuff is for parameters I also do this kind of thing: to keep the names from getting too long. #{input_1 | basename}.bam The basename removes everything after the last . character. Brad ________________________________ From: galaxy-dev-bounces@lists.bx.psu.edu<mailto:galaxy-dev-bounces@lists.bx.psu.edu> <galaxy-dev-bounces@lists.bx.psu.edu<mailto:galaxy-dev-bounces@lists.bx.psu.edu>> on behalf of Jan Hapala <jan@hapala.cz<mailto:jan@hapala.cz>> Sent: Wednesday, November 12, 2014 7:49 AM To: galaxy-dev Subject: [galaxy-dev] rename output dataset in workflow - input dataset variable Hello, I want to rename ouput datasets in a workflow in such a way that the name contains the name of the input dataset. I could not find instructions in the manual (and this info is missing in the Workflow editor - would be really helpful there!). I tried #{input}, e.g. "stats on: #{input}" (with no quotes), but I keep getting just "stats on:" The variable is empty. When I use ${input}, I get a variable field in the workflow form (I do not want this). My G. instance: Galaxy changeset: d1e1beeb532239250396dd8fbdc0156508c6744e<https://bitbucket.org/galaxy/galaxy-central/commits/d1e1beeb532239250396dd8fbdc0156508c6744e> Can anyone help me, please? Jan ___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at: http://lists.bx.psu.edu/ To search Galaxy mailing lists use the unified search at: http://galaxyproject.org/search/mailinglists/
