Extract only sequences that have N nucleotides

Question

[waqasnayab]

Hi,

I have a fasta file. From that fasta file, I need to extract only those sequences that have Ns nucleotides.

>seq1
AGCGGCGTAACGTCGTAGTC
>seq2
ACGCGTACNNNNNNTGCGA

I want output like this:

>seq1
AGCGGCGTAACGTCGTAGTC

Regards,
Waqas

[Devon Ryan]

Your example has no Ns, which is the exact opposite of what you say you want.

Are the sequences always on a single line?

[Sam]

You mean, you want to exclude the sequences with Ns?

awk -v header="" '{ if($1~/^>/){header=$1}else if($1!~/N/){print header; print $1;}}' fastaFile

Note, this only work if you don't have multi-line fasta

Answer 1

[Pierre Lindenbaum]

linearize the fasta

Linearize a fasta sequence

awk -f linearizefasta.awk < input.fa

or

awk '/^>/ {printf("%s%s\t",(N>0?"\n":""),$0);N++;next;} {printf("%s",$0);} END {printf("\n");}' < input.fa

Format back to fasta

tr "\t" "\n" < linearized.tsv

if you know your fasta header have a length < 60

tr "\t" "\n" < linearized.tsv | fold -w 60

view raw README.md hosted with ❤ by GitHub

	/^>/ {printf("%s%s\t",(N>0?"\n":""),$0);N++;next;}
	{printf("%s",$0);}
	END {printf("\n");}

view raw linearizefasta.awk hosted with ❤ by GitHub

filter with awk and a regular expression, convert back to fasta

awk -f linearizefasta.awk < input.fa | awk -F '\t' '($2 ~ /N/)' | tr "\t" "\n"

Answer 2

[Daniel]

With one-line fasta and with sequence names specifically 'seq#' (i.e. no letter 'N's), you can super easily do:

grep -B 1 'N' input.fasta >output_no_Ns.fasta

But that is assuming your data is in the specific format above. Multi line fasta or N's in your headers would break it. But this is what I'd do if I had this data.

Answer 3

[waqasnayab]

Hi,

First, I linearized multi-line fasta file to single-line fasta file:

Multiline Fasta To Single Line Fasta

awk '/^>/ {printf("\n%s\n",$0);next; } { printf("%s",$0);}  END {printf("\n");}' < file.fa

Than, @Daniel command:

grep -B 1 'N' input.fasta >output_no_Ns.fasta

I apologized the community for not explaining my question properly.

Best,
Waqas.

[Daniel]

Thanks for letting us know! You can upvote and tick the helping answers, to finish the question thread.

Answer 4

[Anima Mundi]

In Python, for a file named foo.fa containing linearised FASTAs (prunes FASTAs with Ns):

header = ''

for line in open('foo.fa'):
    if '>' in line:
        header = line
    elif line == '\n':
        pass
    elif 'N' not in line.upper():
        print header,
        print line,

Answer 5

[Brian Bushnell]

With the BBMap package:

reformat.sh in=file.fasta out=fixed.fasta maxns=0