GPU based High Throughput Sequence Mapping for Re-Sequencing Short Reads
The throughput of next-gen sequencers is about 20 to 90 million base pairs per hour and increasing. To map this huge volume of data to reference genome and reduce the computation time, current mapping tools are installed on the clusters. Although using a cluster reduces the computation time but the cost of having such a cluster is considerable. So, there is a trade-off between computation time and computation cost. This project’s goal is to reduce the computation time as well as to reduce the computation cost. The task of mapping the reads is to map all the reads to the same reference genome and it can be parallelized by mapping more than one read at a time. Graphics Processor Units seems to be a promising technology. In this work, we will implement our mapping algorithm on the GPUs. The output of these mappings can be used to assemble the genome or finding the structural variations.