ARTFEED — Contemporary Art Intelligence

PhenixCraft Automates Cryo-EM Model Building with AlphaFold

ai-technology · 2026-05-09

A new fully automated pipeline, PhenixCraft, integrates AlphaFold predictions into the Phenix software to enhance cryo-EM density map segmentation for atomic model building. The method addresses noise and artifacts that typically hinder segmentation, achieving superior TM-scores and sequence accuracy compared to traditional approaches. The pipeline streamlines model building from cryo-EM data, reducing manual intervention and improving efficiency. The work is described in a preprint on arXiv (ID 2605.05259) under Quantitative Biology > Biomolecules.

Key facts

  • PhenixCraft is a fully automated pipeline for building atomic models from cryo-EM density maps.
  • It integrates AlphaFold predictions to enhance map segmentation in Phenix.
  • The method addresses challenges posed by noise and artifacts in cryo-EM data.
  • PhenixCraft demonstrates superior performance in TM-scores and sequence accuracy.
  • It improves upon limitations and inefficiencies of traditional model building using Phenix.
  • The preprint is available on arXiv with ID 2605.05259.
  • The paper is categorized under Quantitative Biology > Biomolecules.
  • The work involves integration of AI predictions into structural biology software.

Entities

Institutions

  • Phenix
  • AlphaFold
  • arXiv

Sources