Creating New Directories While Downloading

[jdduprey]

While downloading the sequence and protein I got some "no such directory exists" errors cause I made a mistake when creating my project directory structure. Is there a way to create a new directory automatically when downloading the file !curl <source file> > ../<some new directory>. Or is there a good reason to avoid doing this?

[jdduprey]

Some additional questions that came up for me:

• Is there a way to use tab completion in jupyter?
• Could we cover the importance of the step where we append the version number to the blast database?
• Is there a way to clearly see how the blast process is progressing/run it in verbose mode?
• Does zipping/unzipping files affect the checksums?
• When is the appropriate time to create/edit the .gitignore file? I couldn't get my github desktop app to not attempt to upload the large sequence files until I deleted the files themselves from the local repository.

[kubu4]

Just documenting discussions from class today:

Is there a way to create a new directory automatically when downloading the file

Possibly (I'm not familiar with most of the curl commands), but it's probably best practice to establish a directory structure before downloading the files.

Is there a way to use tab completion in jupyter?

It might take some time to function after starting a notebook. However, you can force it to happen more quickly by pasting the following line (just the highlighted code part, not the bullet point) in a cell at the top of your notebook and running that cell:

• %config Completer.use_jedi = False

Could we cover the importance of the step where we append the version number to the blast database?

The "version" info is very similar to documenting when a BLAST database file was downloaded. Potentially needed for reproducibility.

Is there a way to clearly see how the blast process is progressing/run it in verbose mode?

Looking at CPU activity (for Macs, the Activity Monitor is useful), frequently checking the line count of the output file (wc -l outfmt6.tab), monitoring the progress of the output file (tail -f outfmt6.tab), and/or checking the last time the output file was modified (ls -l outfmt6.tab).

Does zipping/unzipping files affect the checksums?

The compressed "version" of a file is considered a different file and will have a different checksum than the uncompressed, original file.

When is the appropriate time to create/edit the .gitignore file?

The .gitignore file needs to be edited before you add a file you might want (need?) to ignore. Hopefully next week's readings on git and GitHub will address this.