Received on February 20, 2009; revised on May 6, 2009; accepted on May 12, 2009 | Heng Li and Richard Durbin
The article introduces BWA (Burrows-Wheeler Alignment tool), a new read alignment package that efficiently aligns short sequencing reads against large reference sequences, such as the human genome. BWA supports both base space reads from Illumina machines and color space reads from AB SOLID machines, allowing for gapped alignment and mismatches. The authors compare BWA with MAQ, another hash table-based method, and find that BWA is approximately 10-20 times faster while achieving similar accuracy. BWA outputs alignments in the SAM (Sequence Alignment/Map) format, which can be further processed using the SAMtools software package for variant calling and other downstream analyses. The article also discusses the implementation details of BWA, including the use of the Burrows-Wheeler Transform (BWT) and backward search for efficient alignment. Evaluations on simulated and real data show that BWA performs well in terms of speed, accuracy, and memory usage, making it a promising tool for large-scale short read alignment.The article introduces BWA (Burrows-Wheeler Alignment tool), a new read alignment package that efficiently aligns short sequencing reads against large reference sequences, such as the human genome. BWA supports both base space reads from Illumina machines and color space reads from AB SOLID machines, allowing for gapped alignment and mismatches. The authors compare BWA with MAQ, another hash table-based method, and find that BWA is approximately 10-20 times faster while achieving similar accuracy. BWA outputs alignments in the SAM (Sequence Alignment/Map) format, which can be further processed using the SAMtools software package for variant calling and other downstream analyses. The article also discusses the implementation details of BWA, including the use of the Burrows-Wheeler Transform (BWT) and backward search for efficient alignment. Evaluations on simulated and real data show that BWA performs well in terms of speed, accuracy, and memory usage, making it a promising tool for large-scale short read alignment.