Parallel Local Alignment Search Tool (PLAST)
-------------------------------------------

For the latest information about PLAST visit http://www.irisa.fr/symbiose/projects/plast

I. Overview
 
PLAST is a parallel alignment search tool, used compare two banks containing either protein sequences or nucleotide sequences translated. 
PLAST stands for Faster Search Algorithm; PLAST runs 3 to 5 times faster than NCBI-BLAST for processing large amount data. 

This software was developed using version 4.1.1 of the GNU gcc compiler on a Linux.

The software is freely available for download and open source under the CECILL version 2 license.
PLAST is developed at the Symbiose Team at IRISA-INIRA, Rennes, France. 

The following features are currently available:

 Protein vs protein (plastp), protein vs nucleotide (tplastn), nucleotide vs protein (plastx) and nucleotide vs nucleotide searches (tplastx)
 Query filtering using SEG low complexity filters
 Karlin-Altschul alignment statistics
 Composition-based score adjustments for plastp or tplastn
 

II. Download and installing
 
PLAST is available for download as source code or binarie for x86 Linux systems:

plast-1.0-source.tar.gz
    Source code, unix tar/gzip format
plast-1.0-Linux32.tar.gz
    Intel i686 Linux binary, unix tar/gzip format
plast-1.0-Linux64.tar.gz
    Intel x86_64 Linux binary, unix tar/gzip format

Decompress the above files using the following command:

tar -zxvf plast-1.0-source.tar.gz     
        
To compile the software type:

make 

Which will generate the following binarie:   

plastall

To view a complete list of PLAST parameters simple execute:

$ ./plastall

   -p   Program Name [String]
         plastp, tplastn, plastx and tplastx
   -d   Database [File In]
         defautl =
   -i   Query file [File In]
         defautl =
   -o   PLAST report Output File [File Out]
         defautl =  stdout
   -e   Expectation value (E) [Real]
         defautl = 10.0
   -n   Size of neighbourhood peforming ungapped extension [integer]
         defautl = 22
   -s   Ungapped threshold trigger a small gapped extension [integer]
         defautl = 38
   -z   Effective length of the database (use zero for the real size) [real]
         defautl = 0
   -a   Number of processors to use [Integer]
         defautl = 1
   -G   Cost to open a gap [Integer]
         defautl = 11
   -E   Cost to extend a gap [Integer]
         defautl = 1
   -X   X dropoff value for gapped alignment (in bits) (zero invokes default behavior)
        tplastx 0, all others 15 [Integer]
         defautl = 0
   -Z   X dropoff value for final gapped alignment in bits (0 invokes default behavior)
        tplastx 0, all others 25 [Integer]
         defautl = 0
   -F   Filter query sequence [T/F]
         defautl = T
   -M   Matrix
         defautl = BLOSUM62
   -S   Subset seed [File In]
         defautl
   -R   Query strands to search against database (for plastx and tplastx)
        3 is both, 1 is top, 2 is bottom [Integer]
         defautl = 3
   -C   Use composition-based score adjustments as in Bioinformatics 21:902-911 for plastp or tplastn [T/F]
         defautl = T

References

[1] Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ: Basic local alignment search tool. J Mol Biol 1990, 215, 403-410.

[2] Rognes T, Seeberg E: Six-fold speed-up of the Smith-Waterman sequence database searches using parallel processing on common microprocessors. Bioinformatics 2000, 16, 699-706.


-----------	
Contact.
Van Hoa Nguyen (vhnguyen@irisa.fr)
Dominique Lavenier (lavenier@irisa.fr)