UniProt Protein Scraper — Sequences, Genes, Organisms & Function

Scrape UniProtKB protein data at scale — no API key, no login, no proxy required. Search the world's largest curated protein knowledgebase by keyword, gene, organism or taxonomy, or look up specific accessions, and get clean, structured JSON/CSV/Excel back.

Built on the official UniProt REST API with cursor-based pagination, so results are complete, stable and reproducible.

---

What it does

Give it a list of search queries (insulin, gene:BRCA1 AND organism_id:9606, reviewed:true AND kinase) or a list of accessions (P01308, P0DTC2), and the actor returns one row per protein with:

• accession — primary UniProtKB accession (e.g. P01308)
• entry_name — UniProtKB ID (e.g. INS_HUMAN)
• protein_name — recommended (or submitted) full protein name
• genes — gene names and synonyms
• organism + taxon_id — scientific name and NCBI taxonomy ID
• sequence — full amino-acid sequence
• sequence_length — residue count
• molecular_weight — in Daltons
• function — curated FUNCTION annotation text
• keywords — UniProt controlled-vocabulary keywords
• reviewed — true for Swiss-Prot (manually reviewed), false for TrEMBL
• url — direct link to the UniProt entry page
• source, scraped_at — provenance metadata

---

Why use it

• No API key. Nothing to register, nothing to rotate, nothing to expire.
• No proxy needed. Runs clean on datacenter IPs — cheap runs, no residential surcharge.
• Two modes. Bulk discovery via search, precise enrichment via accessions.
• Full UniProt query syntax. Anything you can type into uniprot.org search works here.
• Reviewed vs unreviewed. Filter Swiss-Prot gold-standard entries with reviewed:true.
• Deduplicated. Every accession appears once, even across overlapping queries.

---

Input

Field	Type	Default	Description
mode	select	search	search (queries → proteins) or accessions (IDs → lookup)
queries	array	["insulin", "spike protein"]	UniProtKB search terms, used in search mode
accessions	array	[]	UniProtKB accession IDs, used in accessions mode
maxItems	integer	100	Maximum proteins to scrape (max 1000)

{
  "mode": "search",
  "queries": ["gene:BRCA1 AND organism_id:9606", "reviewed:true AND kinase"],
  "maxItems": 500
}

{
  "mode": "accessions",
  "accessions": ["P01308", "P0DTC2", "P38398"],
  "maxItems": 100
}

---

Output sample

{
  "accession": "P01308",
  "entry_name": "INS_HUMAN",
  "protein_name": "Insulin",
  "genes": ["INS"],
  "organism": "Homo sapiens",
  "taxon_id": 9606,
  "sequence": "MALWMRLLPLLALLALWGPDPAAAFVNQHLCGSHLVEALYLVCGERGFFYTPKTRREAEDLQVGQVELGGGPGAGSLQPLALEGSLQKRGIVEQCCTSICSLYQLENYCN",
  "sequence_length": 110,
  "molecular_weight": 11981,
  "function": "Insulin decreases blood glucose concentration...",
  "keywords": ["3D-structure", "Diabetes mellitus", "Hormone", "..."],
  "reviewed": true,
  "url": "https://www.uniprot.org/uniprotkb/P01308",
  "source": "uniprot",
  "scraped_at": "2026-07-28T11:00:00+00:00"
}

---

Pricing

Roughly $0.50 per 1,000 proteins in Apify platform usage. Runs on 512 MB and no proxy, so cost stays predictable — a 1,000-protein run typically finishes in a couple of minutes.

---

Use cases

• Bioinformatics — build reference FASTA sets and annotation tables for pipelines
• Drug discovery — pull target sequences, functions and disease keywords for a target list
• Protein research — compare orthologs across organisms by taxonomy ID
• Machine-learning datasets — assemble labelled sequence corpora for protein language models
• Database enrichment — join UniProt annotations onto internal gene or target IDs
• Competitive/landscape analysis — map which proteins carry which curated keywords

---

Useful query examples

Goal	Query
Human proteins only	organism_id:9606
Reviewed human kinases	reviewed:true AND organism_id:9606 AND kinase
One gene, one species	gene:BRCA1 AND organism_id:9606
Membrane proteins	keyword:KW-0472
Everything about a family	family:"insulin family"

---

Related actors

• RCSB PDB Scraper — 3D protein structures
• Ensembl Scraper — genes, transcripts, genomes
• ChEMBL Scraper — bioactive molecules and assays
• PubMed Scraper — biomedical literature

---

Notes

Data comes from the public UniProt REST API. UniProt data is released under CC BY 4.0 — please cite UniProt in publications. This actor only reads public data and does not scrape any personal information.