merge sequences with new sequence insertion

utpalmtbi has asked for the wisdom of the Perl Monks concerning the following question:

Hi Monks,

I have a sequence files (input_file.txt) with huge number of contigs such as (contig number does not reflect the order):

>contig number 11

tttgctcggaggggatc

>contig number 23

gaaaacacttccttattatacaggtaaaccgtatttggat

>contig number 3

aaagctcggaggggatcccct

... ..

I want to concatenate the contigs such that the above order is preserved, and also, I want to insert the sequence "nnnnncattccattcattaattaattaatgaatgaatgnnnnn" in each contig boundaries (here are two contig boundaries), such that the final output file would become as follows:


>concatenated contig 
tttgctcggaggggatcnnnnncattccattcattaattaattaatgaatgaatgnnnnngaaaacactt
+ccttattatacaggtaaaccgtatttggatnnnnncattccattcattaattaattaatgaatgaatgn
+nnnnaaagctcggaggggatcccct
[download]

For concatenation purpose, I use

perl -pe "chomp;s/>.+//" input_file.txt >output_file.txt

But I don't know how to insert the sequence in each contig boundary, plz help..

Comment on merge sequences with new sequence insertion Select or Download Code

Replies are listed 'Best First'.
Re: merge sequences with new sequence insertion by hdb (Monsignor) on Nov 26, 2013 at 20:33 UTC
Read all lines into an array, grep the contigs, and then join them: `use strict; use warnings; my @data = grep { !/^(>\|\s)/ } <DATA>; chomp @data; print join "nnnnncattccattcattaattaattaatgaatgaatgnnnnn", @data; __DATA__ >contig number 11 tttgctcggaggggatc >contig number 23 gaaaacacttccttattatacaggtaaaccgtatttggat >contig number 3 aaagctcggaggggatcccct` [download] or a one liner using regexes: `perl -e "undef $/; print join 'nnnnncattccattcattaattaattaatgaatgaatg +nnnnn', <> =~ /^([tcag]+)$/gm;" input_file.txt >output_file.txt` [download]	[reply] [d/l] [select]
Re: merge sequences with new sequence insertion by boftx (Deacon) on Nov 26, 2013 at 20:22 UTC
Not sure if it can be done as a one-liner, but I would slurp the input file into an array, then use `join` to do the concatenation (with the boundary sequence as the join expression.) It helps to remember that the primary goal is to drain the swamp even when you are hip-deep in alligators.	[reply] [d/l]
Re: merge sequences with new sequence insertion by rnaeye (Friar) on Nov 27, 2013 at 02:54 UTC
How about something like this: #!/usr/bin/perl use warnings; use 5.12.4; my $string = "NNNNNCATTCCATTCATTAATTAATTAATGAATGAATGNNNNN"; while (<DATA>) { next if /^>.+/; next if /^$/; chomp; $_ .= $string unless eof; print; } __DATA__ >contig number 11 tttgctcggaggggatc >contig number 23 gaaaacacttccttattatacaggtaaaccgtatttggat >contig number 3 aaagctcggaggggatcccct ########## It prints: tttgctcggaggggatcNNNNNCATTCCATTCATTAATTAATTAATGAATGAATGNNNNNgaaaacactt +ccttattatacaggtaaaccgtatttggatNNNNNCATTCCATTCATTAATTAATTAATGAATGAATGN +NNNNaaagctcggaggggatcccct [download]	[reply] [d/l]
Re: merge sequences with new sequence insertion by 2teez (Vicar) on Nov 26, 2013 at 21:28 UTC
Hi, You can read one line at a time (for I don't know how large the OP file is), chomp the line, then check if the lines meet your condition using a simply regex, and print it out with the string to insert at the end like so: `use warnings; use strict; my $str = 'nnnnncattccattcattaattaattaatgaatgaatgnnnnn'; while(<DATA>){ chomp; print $_,$str if !/^>\|^$/; } __DATA__ >contig number 11 tttgctcggaggggatc >contig number 23 gaaaacacttccttattatacaggtaaaccgtatttggat >contig number 3 aaagctcggaggggatcccct` [download] This works for me. If you tell me, I'll forget. If you show me, I'll remember. if you involve me, I'll understand. --- Author unknown to me	[reply] [d/l]
Re^2: merge sequences with new sequence insertion by AnomalousMonk (Archbishop) on Nov 26, 2013 at 23:12 UTC
... print it out with the string to insert at the end ... Doesn't that leave you with a `'nnnnnwhatevernnnnn'` "joiner" sequence pasted at the end of the last line of the file? My interpretation of the OP was that joiners should be pasted only between the "contig" sequences. (But I have done no testing of your code.)	[reply] [d/l]
Re^3: merge sequences with new sequence insertion by 2teez (Vicar) on Nov 27, 2013 at 04:59 UTC
Hi AnomalousMonk, Awoshhh!!. My bad! You are right. I think I read wrong, the OP requirement. I think, rnaeye has a better implementation of that. If you tell me, I'll forget. If you show me, I'll remember. if you involve me, I'll understand. --- Author unknown to me	[reply]
Re: merge sequences with new sequence insertion by oiskuu (Hermit) on Nov 26, 2013 at 22:32 UTC
As a one-liner: `perl -ne 'chomp; print "GLUE" x !!$n++, $_ unless /^>/'` [download]	[reply] [d/l]
Re: merge sequences with new sequence insertion by Kenosis (Priest) on Nov 26, 2013 at 20:41 UTC
Perhaps the following will be helpful: `perl -ne 'chomp; $x=0; print if /^(?<!>)\S+$/' input_file.txt >output_ +file.txt` [download] Output on your dataset: `tttgctcggaggggatcgaaaacacttccttattatacaggtaaaccgtatttggataaagctcggaggg +gatcccct` [download] Edit: Sorry... Didn't notice the "insert" spec.	[reply] [d/l] [select]