Script for breaking large .fa files into smaller files of [N] sequences

[lac302]

I'm working on my first WGS assembly submission to NCBI. It's proving to be harder than the assembly itself . NCBI requires assemblies containing more than 20k contigs to be split into chunks of 10k. One of the assemblies I'm submitting has over 800k contigs.

Anyone out there have a script that could handle this? Thanks in advance.

[JackieBadger]

So you essentially have a fasta file with >800k unique sequences?

You can just use the cat or split function in linux

[GenoMax]

http://stackoverflow.com/questions/1...-of-first-line

http://www.biostars.org/p/13270/

There are multiple solutions in the links. You may need to root around to find one that does what you want.

[rnaeye]

Hi!,
You can use script at http://stackoverflow.com/questions/1...nto-many-files

I modified the script little bit and pasted below. If you want bigger chunks, just change "my $record_per_file = 4" value to any chunk size you like.

#!/usr/bin/perl
use warnings;
use 5.12.4;
use File::Basename;
#Split FASTA files into chunks determined by user.
#by RNAeye

my $file = "input.fa"; 		#enter name of your FASTA file here
my $record_per_file = 4;	#Enter how many record you want per file / chunk size
my $file_number = 1;		#this is going to be a part of your new file names.
my $counter = 0;			#counts number of records

open (FASTA,  "<", "$file" )   or die "Cannot open file $file $!";

while (<FASTA>) {
	if (/^>/) { 
		if ($counter++ % $record_per_file == 0) {
			my $basename = basename($file);
			my $new_file_name = $basename. $file_number++ . ".fa"; 
			close(NEW_FASTA);
			open(NEW_FASTA, ">", $new_file_name) or die "Cannot open file $new_file_name $!";
		}
	}
	print NEW_FASTA $_;	
}