I ran deseq2 and then used deg_annotate with a GTF to add annotations to the tabular output of deseq2.
The deseq2 output in galaxy shows headers
When I download the file the headers are gone. Looks like galaxy adds these headers when displaying the data only?
The headers are completely gone after deg_annotate.
The tutorial asks to add headers manually which is very cumbersome for new user
Is there a way the headers can added by deg_annotate and also kept in the downloaded file. The downloaded file is what people usually share use locally.
Thanks
With that context, for your questions. Or, scroll down to the bottom to see my recommendation!
I am not seeing any headers or column numbers in the default view for the tabular files.
This is a generic datatype for “tab separated values”, so column counts 1, 2, .. N are the only “value” and can interfere with data processing if included as a printed line.
I understand editing a workflow starts to become advanced! Scientists will be usually working without headers during processing, then adding custom headers later if wanted. But you could prepare this for the students?
Yes! You could create a customized workflow for your training that includes steps for adding a header baked in?
Warning that this can make the custom workflow somewhat “brittle”. A change in the annotation tool’s output columns would need to be coupled with the Add line to file tool’s text content. Maybe put an annotated box in the workflow canvas with a note?
And I would suggest standardizing the format of the header by running the output through the Convert delimiters to TAB tool (make sure all whitespace is reduced to a single tab!).
For students, adding in the header directly avoids introducing workflow editing too soon.
For “production” use, this can be avoided entirely with workflow customization but it does introduce a dependency.
For Galaxy use, the tool wrapper itself could potentially be updated to add in an optional header based on the advanced settings (use the attributes to label the variable columns) but this would produce a file that will likely need to have the header stripped out to avoid errors, or at least for some use cases.
I know this isn’t exactly what you wanted! Please let me know your thoughts. We can make the request to the tool developers to see what they think. Or you can start an issue up, share it back here, and link in this topic for context in the ticket itself.
Warning that this can make the custom workflow somewhat “brittle”. A change in the annotation tool’s output columns would need to be coupled with the Add line to file tool’s text content. Maybe put an annotated box in the workflow canvas with a note?
