A high performance parallel version of nucleotide BLAST+ for sequence search
#Software Requirements
Any version of MPI (Message Passing Interface) library, either OpenMPI or Intel MPI. OpenMPI is free/open source MPI library and is available at http://www.open-mpi.org
#Download
On Linux:
wget https://github.com/Bioinfo-Tools/MPI-blastn/archive/master.zip -O MPI-blastn.zip
unzip MPI-blastn.zip
On Mac OSX:
curl -L https://github.com/Bioinfo-Tools/MPI-blastn/archive/master.zip -o MPI-blastn.zip
unzip MPI-blastn.zip
#Install
This step requires the mpicc compiler from any MPI library version (OpenMPI, IntelMPI, etc.). OpenMPI is free/open source MPI library and is available at http://www.open-mpi.org. If you are in a cluster or HPC environment, this library may be already installed and you just need to load the module, like in the following example (this command may vary depending on which HPC environment you are using):
module load gcc
module load openmpi
Than you are ready to install:
cd MPI-blastn-master
sh install.sh
If the program was compiled correctly, you will be able to see the file "mpiblastn" inside the "Release" directory.
#Run
After install, MPI-blast can be executed from the Release directory:
mpirun -np <NUM_NODES> ./mpiblastn <INPUT_QUERY> <INPUT_DB> <OUTPUT_TXT> <NUM_CORES>
where:
NUM_NODES is the number of computing nodes
INPUT_QUERY is the name of the input query sequence file in fasta format (http://en.wikipedia.org/wiki/FASTA_format)
INPUT_DB is the name of the input database in BLAST database format (http://selab.janelia.org/people/farrarm/blastdbfmtv4/blastdbfmt.html)
OUTPUT_TXT is the name of the output file (text in tabular format)
NUM_CORES is the number of cores per node to be used
#Examples
The input files listed bellow are available in the examples directory of this software package.
mpirun -np 8 ./mpiblastn Entrada30.fasta NCBI-ntnr_18S_all output.txt 24
In this example above we are running MPI-blastn for the input query sequences in Entrada30.fasta file against the NCBI-ntnr_18S_all database, using 8 nodes and 24 cores/processors per node.
The final result generated in the ouput.txt file is a text file in tabular format, as follows:
Query Subject Identity(%) Matches Gaps Start End E-value Bitscore
Seq_1 gi|1708|emb|X00640.1|Rabbit_18S_ribosomal_RNA_sequence 100.00 1858 0 1 1858 0.0 3333
Seq_2 gi|2485|emb|X58052.1|D.bruxellensis_gene_for_18S_ribosomal_RNA 100.00 1799 0 1 1799 0.0 3245
Seq_3 gi|2588|emb|X58571.1|C.immitus_DNA_for_18S_ribosomal_RNA,_partial 100.00 1713 0 1 1713 0.0 3090
Seq_4 gi|2686|emb|X58053.1|D.hansenii_gene_for_18S_ribosomal_RNA 100.00 1800 0 1 1800 0.0 3247