GPU-based DNA sequence alignment program using Smith-Waterman
Swift is a DNA sequence alignment program that produces gapped alignment using the Smith-Waterman algorithm. It takes in a query file (FASTA format) and a reference file (FASTA format) as input. It outputs the reference name, read name, gapped alignment, alignment score, alignment start and end positions, and alignment length.
I gave a talk on Swift in the GPU Technology Conference 2012. The talk can be accessed at http://www.gputechconf.com/gtcnew/on-demand-GTC.php?sessionTopic=58&searchByKeyword=&submit=&select=+&sessionEvent=&sessionYear=&sessionFormat=#1303.
To install and run Swift, please refer to the Wiki page: http://sourceforge.net/p/swiftseqaligner/wiki/Home/
If you need further help installing or running Swift, please contact Pankaj Gupta at firstname.lastname@example.org.
- Produces gapped alignment using Smith-Waterman algorithm
- Smith-Waterman algorithm runs on the GPU (Graphics Processing Unit)
- Accepts a read file (FASTA format) and a reference file (FASTA format) as inputs
- Outputs reference name, read name, gapped alignment, alignment score, alignment start and end positions, and alignment length
- Supports Illumina reads (fixed length)
- Requirements: Linux OS; 4 GB system memory; CUDA capable graphics card; CUDA toolkit 4.2+; 6 GB global memory; 49 KB shared memory
Be the first to post a review of Swift Sequence Alignment Program!