Add capability for inputting full ForenSeq sequences (not run through UAS)

[rnmitchell]

Adding capability to input full ForenSeq sequences which have not been run through the UAS (#14).

This requires:

• Adding CLI (--uas flag) for indicating if sequences have been run through the UAS for the lusstr annotate command

• Adding the number of bases to remove from the 5' and 3' ends of the full sequence to the str_markers.json file in order to produce the UAS region of the sequence.

• Incorporating code into the annot.py script to remove the 5' and 3' bases to create the sequence of the UAS region (for further annotation).

• Create bracketed annotation form of the full sequence.

[rnmitchell]

Oops... didn't mean to try to merge that...

[rnmitchell]

We can probably merge this branch into the master, @standage. The other things I want to do are a bit more complicated and will require more time.

[standage]

Overall looks good, pending some minor comments.

[standage]

I'll merge when the remaining comments are resolved.

[rnmitchell]

Ok, @standage, I think now it's ready! :)

[rnmitchell]

I'm having issues with writing the code for this test and thought maybe @standage you could solve it in 5 seconds... When I run the test it's cutting off the header of the file it's creating (and so the test is failing)... thoughts? If I remove the header from the comparison file, the test passes. It's driving me crazy.

[rnmitchell]

Ok, so I'm seeing that the NamedTemporaryFile() function creates the file immediately (thus lusSTR sees the file exists and doesn't add the header). Is there another way to create a temp file? Or should I just remove the header of the test file?

[standage]

I was hoping there was a way to tell NamedTemporaryFile to just return a filename without creating the file. There is a mktemp() function, but it is deprecated due to its risk for race conditions.

So the simplest solution I could come up with, silly as it sounds, is to delete the file before running lusSTR annotate.

[standage]

The test suite is passing now, w00t!

Do we have any data sets that we can run without the --uas flag, so that the resolve_uas_sequence() code gets invoked?

[rnmitchell]

yes! Just added the test and all pass for me.