{
  "id": 12908183,
  "title": "NomadFlow: unindexed multi-motif scaffolding with independently mobile rigid groups",
  "url": "https://urgent.news/2026/10/08/nomadflow-unindexed-multi-motif-scaffolding-with-independently-mobile",
  "topic": "science",
  "section": "Science",
  "published": "2026-10-08T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.10.01.754115v1?rss=1"
  },
  "original_language": "en",
  "account": "The research paper \"NomadFlow: unindexed multi-motif scaffolding with independently mobile rigid groups\" introduces a new method for protein structure scaffolding. This method allows motifs, such as active sites or binding epitopes, to be presented in their required geometry, even when the position of the motif along the protein sequence is unknown. The novel approach, called NomadFlow, operates on all-atom SE(3) flow matching, where the motif acts as a condition and each group preserves its internal geometry exactly. Unlike traditional methods that pin down the motif's position, NomadFlow allows the groups to move freely relative to one another until a certain time has passed. After this time, the groups act solely as conditions. The authors present a benchmark for this setting, varying the partition granularity and the structural context around the motif. They achieve success on 19 out of 21 tasks. The trained model proposes a diverse range of valid mutual poses, with the native arrangement often among them. The shape of the solution space varies significantly depending on the geometry of the condition, forming a wide cloud, collapsing onto the native pose, or contracting to a low-dimensional family. Therefore, multi-motif methods should be compared based on the space of layouts they propose and the successful part of it they retain, rather than just yield or fold diversity alone. The researchers have made their code and weights available at https://github.com/ipermyakov/nomadflow.",
  "summary": "Motif scaffolding builds a protein around a prescribed structural motif -- an active site, a binding epitope, a metal site -- so that the motif is presented in its required geometry. The task is usually posed with the motif pinned down: its position along the sequence is given, and several rigid fragments inherit their mutual pose from a native structure. Many applications supply neither. We…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}