v2.1.3

• Bug with cpu count was fixed

• Size of REcomp2 was decreased

• megablast and blastn has been added as options for superclusters identification

• More accurate 'other' contigs selection algorithm based on artificial similarity index
  For each contig in superclusters will be calculated Similarity Index SI:

                                                SI = abs(1 - (N / Q * 100))
  

  where N - number of hits for contig with all consensuses in supercluster, Q - maximal query cover per HSP across all HSP of contig. Contig with the smallest SI value will be selected.

v2.1.2-3

• Test data become more lightweight
• Small bugfixes