removing line breaks and adding carriage returns

Question

[James]

Hi there,

Please can anyone help with this.

I am trying to make a fasta file, created in a Linux NGS pipeline more windows friendly.

I have a fasta file with several sequences in it generated by an NGS pipeline.

I'd really like to reformat the file doing several things, I currently do this manually with a text editor and want to speed things up. Im pretty new to Linux and scripts but guess this could be done much more efficiently. my fasta files looks a bit like this for each gene.

>gene-name-iter4\n
somesequence\n
somesequnece\n
somesequence\n
>gene2name-iter4\n
somesequence\n
somesequence\n

At the moment I remove all \n style line breaks then replace all > with \r\n> to put in windows style carriage returns replace all iter4 with iter4\r\n delete the one carriage return that is now before my first sequence, and add a carriage return to the end of the very last line.

Please can someone show me how to make this easier in Linux, I have access to centos virtual machine, but then move the output consensus fasta file from whole genome sequencing to windows computers.

Thank you James

[cpad0112]

before exporting, you can try unix2dos on linux machine.

[James]

Hi, thanks for the reply. Just tried unix2dos unix2dos -l input.fasta

but that seems to change every \n to \r\n and there are \n at the end of several lines per sequence. Any idea if I should be choosing different options? Thanks

[Ram]

Windows currently has support for \n line endings. Plus, did you try man unix2dos?

[Ram]

What have you tried? This is a fairly common problem, have you tried Googling?

[James]

Hi there, I have tried googling, which is where I sort of found sed or tr should be able to help. For removing all \n I have tried sed 's/\n//g' input.fasta > output.fasta but it doesn't seem to actually do anything, output file is created but still has \n linebreaks I also tried tr -d '\n' input.fasta > output.fasta but that gives me an error tr: extra operand

[cpad0112]

if you want to go sed way:

output:

$ sed 's/\\n//g' test.txt 
>gene-name-iter4
somesequence
somesequnece
somesequence
>gene2name-iter4
somesequence
somesequence

input:

$ cat test.txt 
>gene-name-iter4\n
somesequence\n
somesequnece\n
somesequence\n
>gene2name-iter4\n
somesequence\n
somesequence\n

[James]

Hi thanks for the reply This doesn't seem to work for me,. Sed 's/\n//g' input.fasta > output.fasta All the \n are still there. But Sed 's/iter4/iter4\r\n/g' does work for adding carriage returns everywhere there is iter4

[Ram]

Did you man sed? sed has an extended regex option, and unless you know exactly which sed you're using (pro-tip: mac/BSD sed is the absolute worst), you will need to try a couple of flags, notably the -r flag.