UCSC Genome Browser

UCSC Genome Browser

Overview

The UCSC Genome Browser REST API at https://api.genome.ucsc.edu/ provides programmatic access to genome sequences, annotation tracks, and hub data for 100+ assemblies including hg38, mm39, and dm6. The API is free, requires no authentication, and returns JSON. Use it with the requests library to fetch DNA sequences for genomic regions, retrieve track data (genes, repeats, conservation), list available tracks, and query chromosome sizes for genome-scale coordinate arithmetic.

When to Use

• Fetching the reference DNA sequence for any genomic region (e.g., promoter, exon, CRISPR target) across human, mouse, or other assemblies
• Retrieving RefSeq or GENCODE gene structure (exon coordinates, CDS boundaries, strand) for a locus of interest
• Looking up PhyloP or PhastCons conservation scores to assess evolutionary constraint at a variant site
• Listing and querying any of UCSC's 1000+ annotation tracks (repeats, regulatory elements, conservation) for a region
• Getting chromosome sizes for a genome assembly to set up bedtools, pysam, or coverage pipelines
• Accessing public UCSC track hubs (e.g., ENCODE, Roadmap Epigenomics) without downloading data locally
• Use ensembl-database instead when you need Ensembl stable IDs, VEP variant annotation, or cross-species comparative genomics via the Ensembl REST API
• For bulk local queries across millions of regions, use bedtools-genomic-intervals with pre-downloaded UCSC annotation files

Prerequisites

• Python packages: requests, matplotlib (for visualization)
• Data requirements: genomic coordinates (chrom, start, end in 0-based half-open BED format), genome assembly name (e.g., hg38, mm39)
• Environment: internet connection; no authentication required
• Rate limits: no official published limit; add 0.5s delays for batch requests (>100 queries)

pip install requests matplotlib

Quick Start

import requests

BASE = "https://api.genome.ucsc.edu"

def get_sequence(genome, chrom, start, end):
    """Fetch DNA sequence for a genomic region (0-based, half-open)."""
    r = requests.get(f"{BASE}/getData/sequence",
                     params={"genome": genome, "chrom": chrom,
                             "start": start, "end": end})