-
Notifications
You must be signed in to change notification settings - Fork 0
FAQ
Any PRIDE accession (prefix PXD). Accessions from MassIVE (MSV), jPOST (JPST), and iProX (IPX) are accepted but score Unverifiable. PXAudit only has access to the PRIDE REST API.
No. It only queries metadata endpoints. No raw data, result files, or anything else gets downloaded.
Three-stage deterministic pipeline:
-
Extension registry: maps known extensions like
.mzML,.raw,.mzidto a FileClass -
Exact-stem map: catches MaxQuant fixed filenames like
proteinGroups.txt -
Regex patterns: matches tool-specific output names like
report.tsv,psm.tsv
The PRIDE fileCategory tag is used only as a fallback when none of the above match. Compression suffixes (.gz, .zip, .bz2) are stripped before classification so results.mzid.gz classifies the same as results.mzid.
A few common reasons:
- Missing metadata: title, organism, or instrument absent. Result: None.
- No result files: the submission has raw data but nothing processed. Result: Raw.
- Non-standard results: only proprietary search output, no mzIdentML or mzTab. Result: Bronze.
- No SDRF: no experimental-design file. Result: Silver.
- Partial submission: PARTIAL submissions have relaxed requirements but may still lack files.
Check which accessions used an older version of the tier logic:
SELECT accession FROM audit WHERE tier_logic_version != 'v2.0';Then re-run pxaudit check on those accessions to update their scores.
Partially. Once an accession has been audited, its raw API responses are cached under ~/.pxaudit_cache/. Subsequent runs reuse the cache. If the network is down and stale cached data exists, PXAudit falls back to it with a warning.
Default is 7 days. Use --refresh to force a re-fetch regardless of cache age.
@software{ergin_pxaudit_2026,
author = {Ergin, Enes Kemal},
title = {{PXAudit}: A command-line tool for auditing {Proteomics Exchange} study metadata},
year = {2026},
version = {0.3.0},
url = {https://github.com/LangeLab/PXAudit},
license = {MIT},
}Documentation for PXAudit v0.3.0. Pages can be synced to the GitHub Wiki.
Getting started
Concepts
Contributing