Read mapping tool for AB SOLiD data

Seeds for finding candidate mapping positions We use Iedera [2] to design efficient seeds, based on a model that reflects the reading error distribution observed on SOLiD reads: high chance of errors appearing at distances multiple of 5, and increasing error probability towards the end of the read. Iedera can associate to each seed the list of relevant positions on the read to which the search should be restricted, optimized according to the seed patterns. If configured not to provide this list, the positions are established on each read according to read qualities, basically by avoiding to apply a seed if it requires matches on low quality colors. The false positive hits are detected and eliminated by a second filter that performs a fast SIMD bandwith alignment of the read and the reference.