Proteomics troubleshooting

This page helps you diagnose issues querying the PRIDE and EMDB repositories through the proteomics tools: searches that return no results, PXD or EMDB accession lookups that fail, empty file listings, unexpected resolution-filter behavior, missing linked PDB models, failed map downloads, and API timeouts or rate limiting.

Proteomics troubleshooting

PRIDE search returns no results

Symptom: Zero projects found for a keyword search.

Fixes:

• Try broader search terms. PRIDE indexes project titles, descriptions, and metadata. A very specific protein name may not appear in project-level metadata
• Remove organism, instrument, or modification filters to widen the search
• Check spelling. Scientific names must be exact (e.g., "Homo sapiens" not "human")
• Some recent datasets may not yet be publicly released. PRIDE enforces embargo periods until the associated paper is published

PRIDE Dataset Details fails for a PXD accession

Symptom: Error when looking up a specific project accession.

Fixes:

• Verify the accession format: must be PXD followed by 6 or more digits (e.g., PXD000001)
• Check that the project has been publicly released. Pre-publication datasets are not accessible via the API
• Some very old PRIDE accessions (before the PXD system) may use the legacy format. These are not supported by the current API

File listing is empty

Symptom: PRIDE Dataset Details returns file_count: 0 even though the project exists.

Explanation: Some PRIDE projects are "PARTIAL" submissions that include metadata and identifications but not raw data files. The file listing endpoint may also be temporarily unavailable for projects undergoing re-processing.

Fix: Check the submission_type field. COMPLETE submissions always have files. PARTIAL submissions may not. Visit the PRIDE project page directly for the most up-to-date file listing.

EMDB search returns no entries

Symptom: Zero results for an EMDB keyword search.

Fixes:

• Try simpler search terms. EMDB search is based on title and metadata text
• Protein complex names may differ between databases. Try the common name (e.g., "ribosome" instead of "80S ribosomal complex")
• If you asked for a resolution cutoff, it may be too restrictive. Try a looser cutoff or none
• For organism-specific searches, use the full scientific name (e.g., "Homo sapiens" not "human")

EMDB entry not found

Symptom: Error when looking up a specific EMDB ID.

Fixes:

• Check the ID format. Both EMD-1234 and 1234 are accepted
• Very new depositions may not yet appear in the EMDB API. There can be a delay between deposition and public release
• Some EMDB entries are superseded by newer versions. The original ID may redirect or return an error

Resolution filter returns unexpected results

Symptom: Results include entries with resolution worse than the cutoff you asked for.

Explanation: Some EMDB entries have multiple resolution values (e.g., local resolution vs. global resolution). The filter applies to the reported global resolution, which may differ from local resolution values. Additionally, some entries may not have a resolution value recorded, and these may appear in unfiltered results.

Fix: Review the resolution field in each result. Sort by resolution in your downstream analysis for precise filtering.

Linked PDB IDs are empty

Symptom: An EMDB entry has no linked PDB structures.

Explanation: Not all cryo-EM density maps have fitted atomic models. Some entries are maps-only (no deposited coordinates), especially for lower-resolution structures or novel complexes where model building is ongoing.

Fix: This is expected behavior. Check the EMDB entry page for any related publications that may reference unpublished or in-progress models.

Map download fails

Symptom: The FTP download link for a density map is not accessible.

Fixes:

• FTP access may be blocked by corporate firewalls. Try accessing the EMDB entry page directly and use the HTTPS download option
• Very large maps (>1 GB) may timeout. Use a dedicated FTP client rather than a browser
• The map file is gzipped (.map.gz). You will need to decompress it before use

API timeout or rate limiting

Symptom: Requests fail with timeout or 429 (Too Many Requests) errors.

Fixes:

• The EBI APIs have rate limits. Wait a few seconds and retry
• For batch queries, Ivy spaces requests automatically
• PRIDE search can be slower for very broad queries that match thousands of projects. Narrow your search terms
• EMDB queries are generally fast (<5 seconds) but may slow during server maintenance periods

Getting help

If issues persist:

1. Check the error message for specific details
2. Try the same query on the PRIDE web interface or EMDB web interface to confirm data availability
3. Verify your query format matches the expected input
4. Describe the issue to Ivy

Read next