Fast CPU-based DNA exact sequence aligner

Fast alignment of small sequences to a very large sequence has recently been under attention of many researchers, due to the applications in processing the biological sequences and specially mapping the short reads of Next Generation Sequencing to an already assembled reference genome. The MEMOCODE 2012 contest was aimed to design a very efficient exact DNA sequence matching method to map a huge number of short subsequences of the human genome and to optimize the performance of design: the running time and the cost of the required hardware platform. We designed and implemented two separate solutions, one based on Burrows Wheeler Transform (BWT) with some optimizations to the algorithm that improve its speed significantly, the other one which won the normalized section of the contest uses a hash indexing method.