BACKGROUND: Short-read aligners have recently gained a lot of speed by exploiting the massive parallelism of GPU. An uprising alterative to GPU is Intel MIC; supercomputers like Tianhe-2, currently top of TOP500, is built with 48,000 MIC boards to offer ~55 PFLOPS. The CPU-like architecture of MIC allows CPU-based software to be parallelized easily; however, the performance is often inferior to GPU counterparts as an MIC card contains only ~60 cores (while a GPU card typically has over a thousand cores). RESULTS: To better utilize MIC-enabled computers for NGS data analysis, we developed a new short-read aligner MICA that is optimized in view of MIC's limitation and the extra parallelism inside each MIC core. By utilizing the 512-bit vector units in the MIC and implementing a new seeding strategy, experiments on aligning 150 bp paired-end reads show that MICA using one MIC card is 4.9 times faster than BWA-MEM (using 6 cores of a top-end CPU), and slightly faster than SOAP3-dp (using a GPU). Furthermore, MICA's simplicity allows very efficient scale-up when multiple MIC cards are used in a node (3 cards give a 14.1-fold speedup over BWA-MEM). SUMMARY: MICA can be readily used by MIC-enabled supercomputers for production purpose. We have tested MICA on Tianhe-2 with 90 WGS samples (17.47 Tera-bases), which can be aligned in an hour using 400 nodes. MICA has impressive performance even though MIC is only in its initial stage of development. AVAILABILITY AND IMPLEMENTATION: MICA's source code is freely available at http://sourceforge.net/projects/mica-aligner under GPL v3. SUPPLEMENTARY INFORMATION: Supplementary information is available as "Additional File 1". Datasets are available at www.bio8.cs.hku.hk/dataset/mica.

Abstract Summary: SOAP3 is the first short read alignment tool that leverages the multi-processors in a graphic processing unit (GPU) to achieve a drastic improvement in speed. We adapted the compressed full-text index (BWT) used by SOAP2 in view of the advantages and disadvantages of GPU. When tested with millions of Illumina Hiseq 2000 length-100 bp reads, SOAP3 takes < 30 s to align a million read pairs onto the human reference genome and is at least 7.5 and 20 times faster than BWA and Bowtie, respectively. For aligning reads with up to four mismatches, SOAP3 aligns slightly more reads than BWA and Bowtie; this is because SOAP3, unlike BWA and Bowtie, is not heuristic-based and always reports all answers. Availability: SOAP3 is available at: http://www.cs.hku.hk/2bwt-tools/soap3; http://soap.genomics.org.cn/soap3.html. Contact: [email protected], [email protected]

The advancement of sequencing technologies has made it feasible for researchers to consider many high-throughput biological applications. A core step of these applications is to align an enormous amount of short reads to a reference genome. For example, to resequence a human genome, billions of reads of 35 bp are produced in 1-2 weeks, putting a lot of pressure of faster software for alignment. Based on existing indexing and pattern matching technologies, several short read alignment software have been developed recently. Yet this is still strong need to further improve the speed. In this paper, we show a new indexing data structure called bi-directional BWT, which allows us to build the fastest software for aligning short reads. When compared with existing software (Bowtie is the best), our software is at least 3 times faster for finding unique best alignments, and 25 times faster for finding all possible alignments. We believe that bi-directional BWT is an interesting data structure on its own and could be applied to other pattern matching problems.