I posted my question under "How to integrate custom data sets as hits:synData?" but did not get response so I re-posted it here.
I uploaded my reference tarred file into my synapse project then using the Synapse ID in the hints section but getting the "Custom inputs do not match hints" when running validation. I shared the uploaded file with two challenge team members but I am not sure if this is a correct process.
Would you clarify these steps?
Created by cndgf Thank you all. That solved the problems. Looking at the CWL from your project, it seems that there is a large number of inputs (such as align2, output_filename and others) to the workflow that are required to be defined before the workflow can be run. We don't know what the values of those parameters should be, and only provide a fixed set of input parameters as well as the reference files supplied in the hints stanza. CWL will let you add a 'default' field to the inputs, so that we'll know what to pass in. You can see an example in the RSEM example ( https://github.com/Sage-Bionetworks/SMC-RNA-Examples/blob/master/workflow/smcIsoform-rsem-workflow.cwl#L55 ) where each of the parameters are coded with a default value. So the tool step inputs (not the workflow inputs) would define the values of parameters to be passed in, like `output_filename : { default : my_cool_output.txt }`. Note, the thread count can actually be configured based on the [run time](http://www.commonwl.org/v1.0/CommandLineTool.html#Runtime_environment), so you could have `threadCount : { default: $(runtime.cores) }` Hi Thomas,
Validate from latest https://github.com/Sage-Bionetworks/SMC-RNA-Challenge and dream-smcrna-09072016.tar.gz gives me the same "Custom inputs do not match hints"
We've decided to bump the deadline of the first round to give everyone who is having technical difficulties some extra time. (Now next Wednesday: https://www.synapse.org/#!Synapse:syn2813589/wiki/401443 )
Can you provide an example of the inputs stanza of your workflow? Tomorrow 9/16 is end of round #1 so it seems I have to wait for round #2! That's weird. It looks like the behavior of `cwltool -pack` may have changed.
Namely the class name (synData) looks like it got converted into a file path.
We'll follow up with the cwltool authors. Hi Thomas,
After looking at my merged.cwl, I think the problem comes from "smc_rna_submit.py merge". It added my local host path to the synData
therefore "smc_rna_submit.py validate" is complaining. Please take a look and let me know how to correct it.
Thanks
hint in my CWL file
hints:
- class: synData
input: index
entity: syn5987269
merged_file
"hints": [
{
"input": "index",
"class": "file:///mylocalhost_directory/synData",
"entity": "syn5987269"
}
]
Dear cndgf,
Apologies for the delay in response. You want to include your hints for custom data like so:
```
- class: synData
input: index
entity: syn5909383
```
However, this means that in your `inputs` you must have an input for the custom synData:
```
inputs:
index:
type: File
TUMOR_FASTQ_1:
type: File
TUMOR_FASTQ_2:
type: File
```
Notice how both `index` exists not only in the `inputs` but also the `class: synData`. Please kindly let me know if you have any further questions.
Thanks,
Thomas