Any nice scripts for showing contig length by frequency?

[crader]

Hiya,

has anyone got a script that can nicely show contig length against frequency of contigs at that length?

I've got a bunch of transcriptomes from different platforms and I'd like a nice way of presenting the differences. I could do something using wc -m and output to openoffice, but surely there is a more elegant way of doing it...

otherwise, call me lazy and I'll get on with it!

Cheers

[crader]

I found this, created by Sujai Kumar at The GenePool in Edinburgh:

http://www.google.co.uk/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&ved=0CFoQFjAC&url=http%3A%2F%2Fgenepool.bio.ed.ac.uk%2FGene_Pool%2F454_software%2FScripts%2Fmy_process_contigs.pl&ei=PvOwT5OgCsm48gPPhKCSCQ&usg=AFQjCNHNF_6C6nqWJrPoBWjJAOYtpyF6HA&sig2=8M7xthNZDpDOY3gkMVuMKg

[SES]

Here are some other solutions on this thread: http://seqanswers.com/forums/showthread.php?t=15856

[kmcarr]

Since SES pointed to the script I had posted before I thought I should provide the updated version of it. The version I posted in the previous thread was tailored for getting distributions of miRNAs; this version is more generalized. It will also accept either FASTA or FASTQ input. The script is written in Perl and requires modules Bio::SeqIO (from BioPerl) and Statistics:escriptive::Full. Command line options:

-i|--input   input file name (if none the script will accept input from STDIN)
-f|--format  sequence file format <fasta|fastq> (default fasta)
-s|--start   smallest contig length bin for distribution <n> (integer, default 10)
-e|--end     largest contig length bin for distribution <n> (integer, default 1000)
-b|--bin     size of each bin for distribution <n> (integer, default 20)

Output is some summary statistics (total # of sequences, total sequence length, mean, median, etc.) followed by the contig length distribution data suitable for input to Open Office or Excel (or gnuplot if you wish to extend the script).