locality-sensitive hashing to detect long-read overlaps

 The MinHash Alignment Process (MHAP--pronounced MAP) is a
 reference implementation of a probabilistic sequence overlapping
 algorithm. Designed to efficiently detect all overlaps between noisy
 long-read sequence data. It efficiently estimates Jaccard similarity
 by compressing sequences to their representative fingerprints
 composed on min-mers (minimum k-mer).