FANSe3

The key idea of FANSe is the seeding strategy that uses the minimum number of seeds to cover the read (or the segment of the read of interest). Due to this seed-based core, the search space expands proportionally to the refenence genome size and to the read length, theoretically. This is a disadvantage when compared to the BWT-based algorithms, where the search space narrows down exponentially for each step. However, the seed-based core offers the possibility of mathematical evaluation of mismapping rate - and the mismapping rate is extremely low. Therefore, this seed-based core guaranteed the accuracy and error tolerance. On this basis, we optimized thoroughly by software engineering to achieve a higher speed than the BWT-based methods, without compromising the accuracy.

Based on this concept, a series of hyper-accurate algorithms, not restricted to mapping, were developed, to meed specific demands of various NGS applications.