{
 "page": "tools/index.html",
 "title": "Tool cards",
 "source": {
  "repo": "seanome/2024-kmerseek-analysis",
  "commit": "4a4b11714ddb740f8a7e324058b26bee8fb35af4",
  "commit_date": "2026-10-15T13:12:56Z",
  "path": "results/leaderboard",
  "copied_at": "2026-10-15T13:44:05Z",
  "sha256": {
   "results.parquet": "0c0cf94b5fc40587cab2ffbf7aaf84df9b5475037cc0d818ad2f452a9c73a837",
   "compute.parquet": "8a47ba2123c60cbb41b4833c9f96661934369137b72c5d81ef0808c98247e435",
   "coverage.md": "c10b26730d8dc44b16120bd2f6ee87eb8ed0db3efa40398b339ddac0b0cb7781",
   "provenance.json": "22e54f206354e42208f8a78c4f4827f980ac1db2ce13153f2b077274552e4141",
   "inventory.md": "8dc7040a38cf06af126bd08211edf1f6055c69fe9ac2b4088c6ba8f7f90e6c71",
   "pileup/regions.parquet": "e97701d2fcb601babe25c7d5c6c82760ea19e460e9bc1d555fe5f8f69cb4d6d8",
   "pileup/calls.parquet": "41bc1f6db7bf54f00370b11db2ffed61031d90a42895aa1697be26fc449ea958",
   "pileup/provenance.json": "5ed6e70dd9e82b5200f089b9da3afc62d0a10cc6a51da86715f5f26d3aac9b9e"
  },
  "pileup_commit": "39d81ec90815cacbf3744199d1fa8e99ab28c6f4"
 },
 "text": [
  {
   "tag": "h1",
   "text": "Tool cards"
  },
  {
   "tag": "p",
   "text": "One card per tool, best Fmax first: what it compares, whether it will work for your proteins, where it is strong and weak, what it costs, and the exact command that was run."
  },
  {
   "tag": "p",
   "text": "Data from 15 Oct 2026, commit 4a4b1171 (source)"
  },
  {
   "tag": "summary",
   "text": "Filter the cards"
  },
  {
   "tag": "figcaption",
   "text": "Figure 1. Each tool's best Fmax at each of the 15 relatives, one bar per relative from the nearest (chimpanzee, 6 Mya) to the farthest (E. coli, 2,000 Mya): Pfam domains, raw calls, all proteins, best over the tool's settings. Every card uses the same scale, 0 to 0.45. Hover a bar for its relative and number."
  },
  {
   "tag": "h2",
   "text": "phmmer"
  },
  {
   "tag": "p",
   "text": "Amino-acid letters, one query protein against each protein of the relative, scored with a substitution matrix the way BLAST does but with HMMER's model of insertions and deletions."
  },
  {
   "tag": "li",
   "text": "0.333 best Fmax$0.0186 per 1,000 queriesPfam, raw calls, zebrafish; default options permissive licence"
  },
  {
   "tag": "h2",
   "text": "MMseqs2"
  },
  {
   "tag": "p",
   "text": "Amino-acid letters: it first keeps target proteins that share similar short words with the query, then aligns only those, which makes it much faster than aligning every pair."
  },
  {
   "tag": "li",
   "text": "0.327 best Fmax$0.00113 per 1,000 queriesPfam, raw calls, zebrafish; sensitivity 7 permissive licence"
  },
  {
   "tag": "h2",
   "text": "MMseqs2, three rounds"
  },
  {
   "tag": "p",
   "text": "A profile of a protein family, built by MMseqs2 from its own hits over three rounds of search, as jackhmmer does."
  },
  {
   "tag": "li",
   "text": "0.324 best Fmax$0.00351 per 1,000 queriesPfam, raw calls, zebrafish; sensitivity 7, 3 rounds permissive licence"
  },
  {
   "tag": "h2",
   "text": "HH-suite (hhsearch)"
  },
  {
   "tag": "p",
   "text": "Two profiles against each other."
  },
  {
   "tag": "li",
   "text": "0.313 best Fmax$1.11 per 1,000 queriesPfam, raw calls, zebrafish; single-protein profiles GPL-3.0"
  },
  {
   "tag": "h2",
   "text": "jackhmmer"
  },
  {
   "tag": "p",
   "text": "A profile of a protein family: it searches with one protein, builds a profile from the hits, and searches again with that profile, three rounds in all."
  },
  {
   "tag": "li",
   "text": "0.290 best Fmax$0.109 per 1,000 queriesPfam, raw calls, zebrafish; 3 rounds permissive licence"
  },
  {
   "tag": "h2",
   "text": "Foldseek"
  },
  {
   "tag": "p",
   "text": "Predicted 3-D shapes."
  },
  {
   "tag": "li",
   "text": "0.250 best Fmax$0.00574 per 1,000 queriesPfam, raw calls, zebrafish; shape letters + amino acids needs 3D structure predictionGPL-3.0"
  },
  {
   "tag": "h2",
   "text": "ProstT5 + Foldseek"
  },
  {
   "tag": "p",
   "text": "Shape letters (3Di) predicted from the amino-acid sequence by the ProstT5 protein language model, then searched with Foldseek, so no AlphaFold model is needed."
  },
  {
   "tag": "li",
   "text": "0.223 best Fmax$0.00486 per 1,000 queriesPfam, raw calls, zebrafish; shape letters from sequence needs GPUGPL-3.0"
  },
  {
   "tag": "h2",
   "text": "kmerseek"
  },
  {
   "tag": "p",
   "text": "Short words of k letters (k-mers) shared between two proteins."
  },
  {
   "tag": "li",
   "text": "0.192 best Fmax$0.00246 per 1,000 queriesPfam, raw calls, zebrafish; gbmr4, k = 21 permissive licenceclaims: 1 holds, 2 disproved"
  },
  {
   "tag": "h2",
   "text": "hmmscan with Pfam"
  },
  {
   "tag": "p",
   "text": "Each query protein against the curated profile models of every Pfam family, without searching any relative."
  },
  {
   "tag": "p",
   "text": "Searches Pfam's models, not a relative, so it has one number and no bar per relative."
  },
  {
   "tag": "li",
   "text": "0.886 best Fmax$0.0173 per 1,000 queriesPfam, raw calls, no relative; Pfam models permissive licencegraded on its own models under Pfam"
  }
 ],
 "data": {
  "cards": [
   "foldseek.json",
   "hhblits.json",
   "hmmer3_jackhmmer.json",
   "hmmer3_phmmer.json",
   "hmmscan.json",
   "kmerseek.json",
   "mmseqs2_iterative.json",
   "mmseqs2_seqseq.json",
   "prostt5.json"
  ]
 }
}