[Biojava-l] reading nib sequence files

[mark.schreiber at group.novartis.com]

Would make an interesting comp sci project as you would probably need a 
way to split the DB over multiple files when you hit the limits of your 
OS.

- Mark





Thomas Down <td2 at sanger.ac.uk>
01/24/2005 05:22 PM

 
        To:     Mark Schreiber/GP/Novartis at PH
        cc:     biojava-list List <biojava-l at biojava.org>
        Subject:        Re: [Biojava-l] reading nib sequence files



On 24 Jan 2005, at 09:17, mark.schreiber at group.novartis.com wrote:

> BioJava uses (or at least can use) the PackedSymbolList for large
> sequences. It uses an array of longs to represent the packed bits.
>
> There may be some advantage to using a ByteBuffer, hard to know.

The main reason I was thinking for using MappedByteBuffer is that if 
you're accessing a large amount of sequence it won't necessarily all 
get loaded into memory at once.  This could, for example, make random 
access to a multi-gigabase sequence database bearable on a basic 
desktop computer.  Just a thought, not sure how much demand there is 
for this.

            Thomas.