{
  "id": 2741644,
  "title": "De novo Design of Macrocyclic Molecular Glues",
  "url": "https://urgent.news/2026/08/22/de-novo-design-of-macrocyclic-molecular-glues",
  "topic": "science",
  "section": "Science",
  "published": "2026-08-22T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.08.21.746227v1?rss=1"
  },
  "original_language": "en",
  "account": "In a groundbreaking advancement in drug discovery, scientists have unveiled EvoBind-multimer, a novel deep learning framework designed to de novo engineer molecular glues from protein sequences. This revolutionary approach, distinct from structure-based docking, generates small macrocyclic peptides capable of bridging user-defined protein pairs, bypassing the need for any prior interface knowledge or existing ligands.\n\nThe researchers successfully employed this framework to recruit the E3 ligase VHL to two particularly challenging oncoproteins, KRAS and BRD4. Utilizing Live-cell NanoBRET, they demonstrated robust design-induced proximity for both protein pairs. Mechanistic validation revealed that the generated macrocycles form functional VHL-target ternary complexes, enabling Cullin-RING ligase-dependent proteasomal degradation and downstream signalling shutdown.\n\nHowever, the researchers also discovered that the effectiveness of these macrocycles is deeply context-dependent. Ternary complex processing could act as a potent degrader in one patient model, while simultaneously functioning as a stabilizing LOCKTAC in another, driving VHL-dependent target sequestration without turnover. This new approach of de novo designing induced proximity from sequence alone, through the use of EvoBind-multimer, opens up a new avenue for designing novel protein functions.",
  "summary": "The engineering of induced proximity has transformed drug discovery, yet the development of molecular glues remains largely serendipitous and restricted to the retrospective optimisation of accidental discoveries. Here, we present EvoBind-multimer, a deep learning framework for the de novo design of molecular glues directly from protein sequences. Unlike structure-based docking, our method…",
  "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."
}