Output to a file

We can store the result of our analyses in files using the > operator.

For example, to store all burrito orders into burritos.tsv:

Now notice that ls shows the newly-created file!

Let’s take a brief detour from burritos and talk about bioinformatics.

If you’re using grep to extract lines from a FASTA file that contain the > character, make sure to put it in quotes! Otherwise, grep interprets > as an operator and will truncate your FASTA file and your data will be lost (there is no “Trash” when working on the command line).

Let’s illustrate this pitfall in our sandbox. If we grep for > with quotes, our 3 chromosomes are listed as expected:

But, not quoting results in a truncated file!

Using ls, verify that ref.fa is empty (i.e. 0 in the 5th column):

Not to worry, we had prepared a backup file ref.fa.bak that you can use to regenerate ref.fa with the cp (copy) command:

This is a common rite of passage in bioinformatics, so it’s good to get it out of the way in a virtual sandbox!

And although putting quotes around > fixes the problem, a better way to avoid this issue altogether is to always prepend cat to your command-line pipelines (cat returns the contents of a file):

With this approach, missing quotes lead to an error message instead of data loss.