UNIX for Dummies Questions & Answers

So I have a file that contains

Code:

>NM_#########AUGCAUCGUAGCUAGUCGAUACUGGACUG>NM_########AUGAGUAUGUAUGAUGUAUGUAUGA

where # is any digit 0-9 (the text is many repetitions of the pattern above, not just that, but all in one line), and I want it to show

Code:

>NM_#########
AUGUAGUGCUAGCUGAUCGAUGCUAGUCGUAGC
>NM_########
AGUGAGUCGUCGUGACUGACUGUGGCAUCGUA

Basically I need to add a new line before every > and between a number and a letter.

OR

If it's easier, I have something like this

Code:

>NM_#########
AUGCUGAC
GACGUAGC
ACGUGUAG
>NM_########
AGUGCUGA
ACGUAGCU
ACGUGCUA

and I want to condense all the letter only lines to one line, like the output shown above.

If someone could help me how to do this with a simple command, it would really help.

Thank you!

It would be better if you posted your sample input without masking digits with "#"'s... Anyway try this:

Code:

perl -pe 's/(\d)(\D)/$1\n$2/g;s/(\w)>/$1\n>/g' file

continuous the topic, add one more sed command to reformat the output.

Code:

perl -pe 's/(\d)(\D)/$1\n$2/g;s/(\w)>/$1\n>/g' infile |sed '/^[!A-Z]/ s/.\{8\}/&\n/g'


>NM_123455673
AUGCAUCG
UAGCUAGU
CGAUACUG
GACUG
>NM_12345634
AUGAGUAU
GUAUGAUG
UAUGUAUG
A