Skip to content

DIAMOND v2.2.6

Choose a tag to compare

View all tags
@bbuchfink bbuchfink released this 02 Sep 09:00
· 12 commits to master since this release
v2.2.6
f63726e
  • Improved the performance of the search and clustering workflows.
  • Improved the sensitivity of the --very-sensitive alignment mode.
  • Fixed potential issues of increased memory use in the clustering workflows.
  • Added the option --linclust-minichunk to set the minimal chunk size for linear-time clustering rounds.
  • Added limited support for the AVX512 machine instructions and the respective runtime dispatch code.
  • The SSE4.1 compatible architecture code path now also requires SSE4.2, CMPXCHG16B and SAHF.
  • The AVX2 compatible architecture code path now also requires BMI, BMI2, LZCNT, F16C, MOVBE, AES, PCLMUL, XSAVE, XRSTOR and XSAVEOPT.
  • Added detection of the loong64 architecture to CMakeLists.txt.
  • Fixed potential compiler errors on the loong64, ppc64el, s390x, ppc64 and sparc64 architectures.
  • Added the countdistinct workflow to count distinct sequences in a FASTA file.
  • Added the option --no-reorder to disable reordering of alignment output according to the order of queries in the input file.
  • Added the option --no-mempool to disable per-thread memory pools.
Assets 5
Loading