Provided by: mhap_2.1.3+dfsg-2_all 

NAME
mhap - probabilistic sequence overlapping
SYNOPSIS
mhap -s<fasta/dat from/self file> [-q<fasta/dat to file>] [-f<kmer filter list, must be sorted>] java -server -Xmx<memory> -jar $(which mhap) -s<fasta/dat from/self file> [-q<fasta/dat to file>] [-f<kmer filter list, must be sorted>] generate precomputed binaries: java -server -Xmx<memory> -jar $(which mhap) -p<directory of fasta files> -q <output directory> [-f<kmer filter list, must be sorted>]
DESCRIPTION
MHAP (MinHash Alignment Protocol) is a tool for finding overlaps of long-read sequences (such as PacBio or Nanopore) in bioinformatics.
OPTIONS
-h --help Show summary of available options. --version Displays the version.