pride-database

What it does

The skill enables programmatic exploration of the PRIDE Archive v3 REST API. It supports searching for projects by free-text keyword and facet filters (organism, instrument, disease, software), retrieving full project metadata, and listing or downloading associated files such as RAW, PEAK, RESULT, and FASTA. It also provides a way to look up which PRIDE projects mention a given UniProt accession and to find similar projects. For proteins and domains, it directs usage of external databases (uniprot-protein-database for sequences, interpro-database for domain architecture).

How it works

• Describes usage of PRIDE v3 endpoints: /search/projects for keyword + facet filtering, /facet/projects to discover facet values, /projects/{accession} for full metadata, /projects/{accession}/files or /files/all for project files, /files/sdrf/{accession} for SDRF, /proteins/{accession} for project mappings, and /projects/{accession}/similarProjects for similarity data. It shows example Python snippets that perform HTTP GET requests to these endpoints, parse JSON responses, and extract fields such as accession, title, submissionDate, instruments, organisms, DOIs, and file lists with their categories and public locations. It emphasizes the absence of peptide/PSM endpoints in PRIDE v3 and instructs to use RESULT files for spectrum-level data.
• It instructs how to paginate through results, how to filter on file categories (RAW, PEAK, RESULT, FASTA, OTHER), and how to compute aggregate file sizes from fileSizeBytes fields. It includes code that prints summaries like top projects, file counts by category, and SDRF availability.
• It includes explicit commands and file names like "/search/projects", "/facet/projects", "/projects/{accession}", "/projects/{accession}/files", "/files/all", "/files/sdrf/{projectAccession}", "/proteins/{accession}", "/projects/{accession}/similarProjects", and calls to external databases (uniprot-protein-database, interpro-database).

When to use it

• When you need to find published proteomics datasets by keyword and facet filters (organism, tissue, disease, instrument, software, PTM) for meta-analysis or benchmarking.
• When you want to download raw mass-spectrometry data (RAW, mzML, MGF) or pre-processed identifications (RESULT files) from a specific PRIDE project.
• When you need to determine which PRIDE projects mention a specific UniProt accession (project-level occurrence).
• When you want to discover similar projects or obtain autocomplete suggestions to guide searches.
• When you need to fetch SDRF mappings to model sample-to-MS-run relationships.

What it can touch

• The PRIDE v3 REST API endpoints as listed above. It demonstrates usage via Python's requests to interact with: /search/projects, /facet/projects, /projects/{accession}, /projects/{accession}/files, /files/all, /files/sdrf/{projectAccession}, /proteins/{accession}, /projects/{accession}/similarProjects, and /search/autocomplete.
• It references external databases for protein sequences and domain annotations: uniprot-protein-database and interpro-database.

Caveats

• PRIDE v3 has no /peptides, /psms, or /proteins?proteinAccession= endpoints; peptide/PSM data are only available in local RESULT file parsing after download.
• Endpoints return plain JSON arrays (no HAL), and some endpoints may require pagination for large results.
• No authentication is required for PRIDE v3 endpoints, but rate-limiting guidance suggests bursts under ~5 requests/second with sleeps in loops.
• The skill assumes internet access and Python package availability (requests, pandas, matplotlib).