{
  "annotation_id": "5b28c1d0-b553-485e-9246-22eea6150c7d",
  "date_created": "2026-03-02T18:01:31.871000Z",
  "date_modified": "2026-03-02T18:01:31.871000Z",
  "file_hash": "77e56b5f078fdde56338548e11490c5d9d5f4f3427545689ecbd70b099a723f2",
  "private": false,
  "record": {
    "abstract": "Circular permutation connects the N and C termini of a protein and\nconcurrently cleaves elsewhere in the chain, providing an important mechanism\nfor generating novel protein fold and functions. However, their in genomes is\nunknown because current detection methods can miss many occurances, mistaking\nrandom repeats as circular permutation. Here we develop a method for detecting\ncircularly permuted proteins from structural comparison. Sequence order\nindependent alignment of protein structures can be regarded as a special case\nof the maximum-weight independent set problem, which is known to be\ncomputationally hard. We develop an efficient approximation algorithm by\nrepeatedly solving relaxations of an appropriate intermediate integer\nprogramming formulation, we show that the approximation ratio is much better\nthen the theoretical worst case ratio of $r = 1/4$. Circularly permuted\nproteins reported in literature can be identified rapidly with our method,\nwhile they escape the detection by publicly available servers for structural\nalignment.",
    "arxiv_id": "q-bio/0407039",
    "authors": [
      "T. Andrew Binkowski",
      "Bhaskar DasGupta",
      "Jie Liang"
    ],
    "categories": [
      "q-bio.BM"
    ],
    "doi": "10.1109/IEMBS.2004.1403795",
    "title": "Order independent structural alignment of circularly permuted proteins",
    "url": "https://arxiv.org/abs/q-bio/0407039"
  },
  "schema_id": "dorsal/arxiv",
  "source": {
    "execution_id": "b48cec83-76c5-4228-a1b8-122350e00056",
    "id": "arXiv Dataset IDs",
    "type": "Model",
    "variant": "snapshot-2026-03-01",
    "version": "0.1.0"
  },
  "user_id": 1000002
}