{
  "id": 18107849,
  "title": "Decision benchmarking of cryo-EM re-refinement",
  "url": "https://urgent.news/2026/11/04/decision-benchmarking-of-cryo-em-re-refinement",
  "topic": "science",
  "section": "Science",
  "published": "2026-11-04T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.10.31.763284v1?rss=1"
  },
  "original_language": "en",
  "account": "Cryo-EM re-refinement often employs map fit, with the deposited model serving as the reference. Researchers have compared the performance of retaining the model (no operation) to using other selection criteria. In a cohort of 167 entries with 131 sequence groups, the selection based on correlation against higher-resolution references resulted in a 2.4-degree increase in chi-square disagreement (95% confidence interval of 2.1-2.7 degrees). Positive loss was observed in 91% of the groups. The group-weighted fractions showed 7.2% and 3.7% increases and decreases in chi-square disagreement, respectively. The fraction of cases exceeding 1 Angstrom in local side-chain RMSD rose by 2.4 percentage points.\n\nSelecting the half map that had not been used in the re-refinement process increased the disagreement by 2.29 degrees (interval of 1.70-2.97 degrees) across 99 entries. In 53 prespecified pairs of low/high-resolution within-studies, the loss in agreement was 3.43 degrees (interval of 2.85-4.02 degrees), with 50 pairs showing positive outcomes. Half-map loss was 2.35 degrees (interval of 1.60-3.21 degrees) in 44 pairs. The Primum software provides fixed cohorts, residue sets, and scoring tools to benchmark refiners and selection rules against a no operation scenario.",
  "summary": "Map fit is widely used in cryo-EM re-refinement. We benchmarked selection against retaining the deposited model (no operation), using packed-side-chain agreement with higher-resolution references. In a prespecified 3.0-3.6 [A] cohort (167 entries, 131 sequence groups), highest-correlation selection increased {chi}1 disagreement by 2.4{degrees} (95% interval 2.1-2.7{degrees}), with positive loss…",
  "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."
}