Bug in greedy lowmem clustering? #664
Comments
martin-steinegger
added a commit
that referenced
this issue
Jun 18, 2023
martin-steinegger
added a commit
to steineggerlab/foldseek
that referenced
this issue
Jul 31, 2023
80f8b0be Add check on seqdb to mkrepseqdb 542f3621 Add mkrepseqdb as command 8459b6b3 disable wrapped scoring when target sequence is shorter f5f780ac ca3m segfault same db 25688290 Update greedy cluster algo. to improve issue reported here: soedinglab/MMseqs2#664 03f0bcca Remove TestIndexTable 07ca4a7c Allow kseq_t to read sequences larger than 2^31 bytes 3e436173 Patches to allow MMseqs2 to build with gcc13 (#714) 71dd32ec Add target similiar k-mer search to prefiler 390457d8 Fix warning in pairaln 15140153 Rework pairaln to support different pairing modes. Add support for dummy sequences to result2msa 12ba2027 Try to fix cirrus 38c5798d Switch azure binary upload to new uploader git-subtree-dir: lib/mmseqs git-subtree-split: 80f8b0bed1e8b005455a4ca81b69d469f8c577b4
elileka
added a commit
to soedinglab/metaeuk
that referenced
this issue
Sep 4, 2023
df77d9e6cf Expose Smith-Waterman-based prefilter 1d62fa0cdd Fix source number being limited to 16-bit (65k) #729 4b52296253 Fix database creation for GTDB r214 (#742) ad6dfc66d7 Changes needed for cluster search eb01b5b764 Correctly expose db-load-mode in ungapped prefilter 8fe3bf9bbe All combining various index subset modes for createindex 91f2a6ac26 Handle _h the same as other DBs in createclusterdb 8310cd6bc4 Reorder header file by clusters too 95e1c1043a Remove the rel path 001902671a Fix sym link issue 76b7df1e3f Fix rel/abs issue in createclusearchdb 9ae4458a5c Add createclusearchdb, rm mkrepseqdb 80f8b0bed1 Add check on seqdb to mkrepseqdb 542f362121 Add mkrepseqdb as command 8459b6b30e disable wrapped scoring when target sequence is shorter f5f780acd6 ca3m segfault same db 25688290f1 Update greedy cluster algo. to improve issue reported here: soedinglab/MMseqs2#664 03f0bcca33 Remove TestIndexTable 07ca4a7c5c Allow kseq_t to read sequences larger than 2^31 bytes 3e43617332 Patches to allow MMseqs2 to build with gcc13 (#714) 71dd32ec43 Add target similiar k-mer search to prefiler 390457d87e Fix warning in pairaln 1514015351 Rework pairaln to support different pairing modes. Add support for dummy sequences to result2msa git-subtree-dir: lib/mmseqs git-subtree-split: df77d9e6cf640fe8990f247441ab44d4f4ad9cdc
gamcil
pushed a commit
to steineggerlab/foldmason
that referenced
this issue
Nov 28, 2023
80f8b0be Add check on seqdb to mkrepseqdb 542f3621 Add mkrepseqdb as command 8459b6b3 disable wrapped scoring when target sequence is shorter f5f780ac ca3m segfault same db 25688290 Update greedy cluster algo. to improve issue reported here: soedinglab/MMseqs2#664 03f0bcca Remove TestIndexTable 07ca4a7c Allow kseq_t to read sequences larger than 2^31 bytes 3e436173 Patches to allow MMseqs2 to build with gcc13 (#714) 71dd32ec Add target similiar k-mer search to prefiler 390457d8 Fix warning in pairaln 15140153 Rework pairaln to support different pairing modes. Add support for dummy sequences to result2msa 12ba2027 Try to fix cirrus 38c5798d Switch azure binary upload to new uploader git-subtree-dir: lib/mmseqs git-subtree-split: 80f8b0bed1e8b005455a4ca81b69d469f8c577b4
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Hi, while benchmarking your tool I came across an issue clustering with
--cov-mode 1. Clusterable pairs remain in the representative set that should be found by the search workflow.Steps to reproduce this using release 14 and the astral95 set (http://scop.berkeley.edu/downloads/scopeseq-2.08/astral-scopedom-seqres-gd-sel-gs-bib-95-2.08.fa):
Running the same command lines using release 11 and
--cluster-mode 2results in 9 hits instead of 1231. Looks like a bug or am I not understanding something?The text was updated successfully, but these errors were encountered: