Experiment-guided AlphaFold3 Predictions for Protein Conformational Ensembles

AlphaFold3 predictions guided by NMR spectroscopy, X-ray crystallography, and cryo-EM experimental data to resolve accurate protein ensembles. The methodology, developed by Alexander Bronstein and hosted on Harvard Dataverse, was last updated on May 4, 2026. It aims to capture the heterogeneous nature of protein structures that traditional predictions may collapse into a single conformation.

Use Cases

• Refining protein structure predictions based on NMR distance restraints mentioned in the description.
• Identifying alternate protein conformations based on unmodelled electron density data mentioned in the description.
• Improving model quality by jointly guiding predictions with cryo-EM maps and NMR parameters as described.
• Developing experimentally aware predictive models that capture structural ensembles as outlined in the description.

Strengths

• Methodology demonstrates reduced distance restraint violations compared to traditionally resolved NMR structures.
• Can uncover unmodelled alternate conformations detectable in experimental electron density.
• Integrates guidance from multiple experimental techniques: NMR, X-ray crystallography, and cryo-EM.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

Harvard Dataverse

Collection Method

Computational predictions from AlphaFold3 guided by experimental data.

Time Range

null

Freshness

Last updated 2026-05-04 22:54:32; freshness should be verified.

Geography

null