Trim sequences in FASTA file from sanger sequencing

Question

[merajazizmeraj]

Is there a tool out there that trims FASTA sequences at the beginning and ends.

By trimming i mean take the input sequences and generate the ouputs. I have more than 3000 sequences.

Thanks.

INPUT:

>Seq1
NNNNNNNNNNTTGNNNNGGATNTCCTTTCCGAATATTTTTGGTGCATTTGTAATAAATGTCATTTNTCTCCTTTTTAAAGGAATTGTCTTAGAAGAAAGAAGGCAAGCCACCATTTTACCCACGTAAATATATGAATATATTTCTGACATTGAGGTGTTCCAGAAGATGATAAAGAAATGATAGCAGCTCCAGAAATACCAACTGATTTTAATCTACTACAGTAAGTAAATTATATTCTGATAATTTTTAAATACTTGTTTATTCCACAAAATGGGGAATGCATTAACTTCAGTTAAATTTCCTTCTGCTCGAGAAGATCTAATATATAAAATAGCTTTTATGCTTTGCAAGAGTTTATATCAGNANCNNNNNNNNNNCNGN
>Seq2
NNNANNNNNGNNNNGTATGANGTTTTGGGGAACATCTTAATTACTTATAATGCTAATATGAAGTTTTGTAATGAGTTAACCAAGCCTTTCTTTTAGAAAATATGGCAAAAATTAGAAACTCAATATAAATTTCTAAGGAAGGGTTTTAATTCTTATCTTTCTGTCACAGGGAGTCAGAAACACATTTTTCTTCTGACACAGATTTTGAAGATATCGAAGGAAAAAACCAAAAGCAAGGCAAAGGCAAAGTATGTATCAAATATTTGACTTTATTTTGTTTCCTAAGATCTCACACACACACAGATTTAAGTTATGTCTCAGATAGTTTTATCTTTTAAAAATGGCTTTTTAAGGGGGTGGGAGCTGATTGGTATGGTAANCAN
>Seq3
NNNNNGNNNNNNNNNTNNNTNNNTNNNAAGTGGATGGAATTCTTTAGGGCAAGTTTAAGCATGTTATGTACCCTATCAGCTACTTCTACTGTAGCTGTGTTTTGAACTCTCAAGGATAGTGATATAACTTAACCACCTCGTATTTTTTATGCAGACTTGTAAAAAAGGCAAAAAGGGCCCAGCAGAAAAGGGCAAAGGTGGAAATGGAGGAGGAAAACCTCCTTCTGGTCCAAACCGAATGAATGGTCATCACCAACAGAATGGAGTGGAAAACATGATGTTGTTTGAAGTTGTTAAAATGGGCAAGAGTGCTATGCAGGTAAGATTTATGTTGTTCTTCCCAGTTCATTTGTACATTTTAAACTTTAATGAGTTATATAGAGTGTAGCTCTGNNNNNNNNNNTTGCAA
>Seq4
NNNNNNNCNCNNNNNGNGNNNNCNAAGTGACTATTTGAGAGCTGCTGATTTCAAAATAAATATATCTTACCTTTACAGCCTGAACACTGAATAAAAAAGTTGATAAGGTCAAGAAGTGCTATATCTCGGTCATGCTTGTATGATTCTATCCAATCATCTACCACCGACTACAGCAGAGGGAAAAAAATAAAATCATTAGCTTCTTCTAATTTTCTCAAAATCAATTAAGTCTGATAAAGTCATAAAATTCAAGATTATATAGTATCACATTACTTTAATATAAATACTTATACACTGAAATTTAAAGTTCAATTTTAACAATAATAAAATAGAATCGAATTCAGTAAAACAATTATCTGATAACACAAAATGACCTATCAATCTTCTATTTATTTTGCATTGAAAAGAATGTGGNNN
>Seq5
NNNNNNNNNAANNNNNNNNNNNNNNNNNNNTNNNNANNNNNNNNNNNTAAGTTATCAAAACACTTAAGGTAGTAAGTTACCTCATCGAATTCTTCAGTCATTTTTCGAATTATCTCAGAGTTCTGCATATGTCTAAACATTTCTGCTGTGACAACTCCTGAAATTTGCAAATGTCAGAAGTTAATATATGGTGTGATAAAAAAATAAAGAAAACTTCCAAGTAAGTCTCTAACACTAAGAAGTCTATGGTCACACAATAAAAGGCATACTTCTTCAACCATCATCTAATAATCTTTACCATGATACTCTAATCTATAAATAAAGCACAAACAAATGCTATCTATTCTCAGTATGCACAAGAAAACAGCCCCATACTTCTGACAGATATCTTTTTTCCTAACACAATTAACTTTGGCCATTTCTANNNNNNNNNNTTNNNNAAN

OUTPUT:

>Seq1
TCCTTTCCGAATATTTTTGGTGCATTTGTAATAAATGTCATTTNTCTCCTTTTTAAAGGAATTGTCTTAGAAGAAAGAAGGCAAGCCACCATTTTACCCACGTAAATATATGAATATATTTCTGACATTGAGGTGTTCCAGAAGATGATAAAGAAATGATAGCAGCTCCAGAAATACCAACTGATTTTAATCTACTACAGTAAGTAAATTATATTCTGATAATTTTTAAATACTTGTTTATTCCACAAAATGGGGAATGCATTAACTTCAGTTAAATTTCCTTCTGCTCGAGAAGATCTAATATATAAAATAGCTTTTATGCTTTGCAAGAGTTTATATCAGNANC
>Seq2
GTATGANGTTTTGGGGAACATCTTAATTACTTATAATGCTAATATGAAGTTTTGTAATGAGTTAACCAAGCCTTTCTTTTAGAAAATATGGCAAAAATTAGAAACTCAATATAAATTTCTAAGGAAGGGTTTTAATTCTTATCTTTCTGTCACAGGGAGTCAGAAACACATTTTTCTTCTGACACAGATTTTGAAGATATCGAAGGAAAAAACCAAAAGCAAGGCAAAGGCAAAGTATGTATCAAATATTTGACTTTATTTTGTTTCCTAAGATCTCACACACACACAGATTTAAGTTATGTCTCAGATAGTTTTATCTTTTAAAAATGGCTTTTTAAGGGGGTGGGAGCTGATTGGTATGGTAANCA
>Seq3
AAGTGGATGGAATTCTTTAGGGCAAGTTTAAGCATGTTATGTACCCTATCAGCTACTTCTACTGTAGCTGTGTTTTGAACTCTCAAGGATAGTGATATAACTTAACCACCTCGTATTTTTTATGCAGACTTGTAAAAAAGGCAAAAAGGGCCCAGCAGAAAAGGGCAAAGGTGGAAATGGAGGAGGAAAACCTCCTTCTGGTCCAAACCGAATGAATGGTCATCACCAACAGAATGGAGTGGAAAACATGATGTTGTTTGAAGTTGTTAAAATGGGCAAGAGTGCTATGCAGGTAAGATTTATGTTGTTCTTCCCAGTTCATTTGTACATTTTAAACTTTAATGAGTTATATAGAGTGTAGCTCTG
>Seq4
AAGTGACTATTTGAGAGCTGCTGATTTCAAAATAAATATATCTTACCTTTACAGCCTGAACACTGAATAAAAAAGTTGATAAGGTCAAGAAGTGCTATATCTCGGTCATGCTTGTATGATTCTATCCAATCATCTACCACCGACTACAGCAGAGGGAAAAAAATAAAATCATTAGCTTCTTCTAATTTTCTCAAAATCAATTAAGTCTGATAAAGTCATAAAATTCAAGATTATATAGTATCACATTACTTTAATATAAATACTTATACACTGAAATTTAAAGTTCAATTTTAACAATAATAAAATAGAATCGAATTCAGTAAAACAATTATCTGATAACACAAAATGACCTATCAATCTTCTATTTATTTTGCATTGAAAAGAATGTGG
>Seq5
TAAGTTATCAAAACACTTAAGGTAGTAAGTTACCTCATCGAATTCTTCAGTCATTTTTCGAATTATCTCAGAGTTCTGCATATGTCTAAACATTTCTGCTGTGACAACTCCTGAAATTTGCAAATGTCAGAAGTTAATATATGGTGTGATAAAAAAATAAAGAAAACTTCCAAGTAAGTCTCTAACACTAAGAAGTCTATGGTCACACAATAAAAGGCATACTTCTTCAACCATCATCTAATAATCTTTACCATGATACTCTAATCTATAAATAAAGCACAAACAAATGCTATCTATTCTCAGTATGCACAAGAAAACAGCCCCATACTTCTGACAGATATCTTTTTTCCTAACACAATTAACTTTGGCCATTTCTA

[Istvan Albert]

just to clarify your post, what you want to happen is to trim back from both the 5' and 3' ends to the point where no more N's are visible within a certain distance

[merajazizmeraj]

yes...that is correct.

Answer 1

[Devon Ryan]

Here's an awk solution (because why not). It looks at bins of 5 bases and will trim them off either end if they contain an N. You can modify this at will, of course. Just change foo.fa to whatever your file is called and then pipe things to a new file.

awk '{
header=$0;
getline;
for(five_prime=1;five_prime<length($1)-5;five_prime++) {
    s=substr($1,five_prime,5);
    if(index(s,"N")==0) break;
}
for(three_prime=length($1)-4;three_prime>five_prime;three_prime--) {
    s=substr($1,three_prime,5);
    if(index(s,"N")==0) break;
}
printf("%s\n%s\n",header,substr($1,five_prime,three_prime-five_prime+5));
}' foo.fa

Edit: Fixed an off-by-one error.

Answer 2

[xb]

How about this,

FASTA/Q Trimmer

[merajazizmeraj]

Thanks. I did look at this and am using galaxy but it is not trimming the beginning and ends. It just removes sequences that have lots of NNNNNNN's in them.

Answer 3

[Istvan Albert]

A more generic solution for your problem could be to find the longest substring that is bounded by Ns. A simple python script like the one below could do that:

import sys
for line in sys.stdin:
   if line[0] == ">":
        print line 
        continue
    line = line.strip()
    pieces = line.split("N")
    sizes = sorted(((len(p), p) for p in pieces), reverse=True)
    longest = sizes[0][1]
    print longest

Run it with python trim.py < input.fasta

[merajazizmeraj]

wont this split the string if there are a few N's in between the sequences. I would like to trim the ends as much as possible

and tolerate some N's in the middle. Thanks

[Istvan Albert]

correct, like I said this will give you the longest substring that is bounded by Ns

it is just a different way to think about the problem, and when one does so they may identify different requirements