The UniProt website API: facilitating programmatic access to protein knowledge

Abstract

The UniProt REST API is a freely available, open-access resource that powers the UniProt.org website and gives users flexible programmatic interaction with protein knowledge data. It provides access to UniProtKB, UniRef, UniParc, Proteomes, GeneCentric, ARBA, UniRule, and the ID Mapping tool, along with supporting data and controlled vocabularies. Users can access the API with their favorite programming language and generate example code snippets to access the UniProt databases using the API documentation page (https://www.uniprot.org/api-documentation) in various languages. API results can be returned and downloaded in various formats. With an average of 303 million requests per month over the last year, the API enables structured search queries using logical operators and parentheses, allows users to specify fields of interest within results, and customize downloads for direct integration into workflows. The API is a powerful tool that empowers users to fully utilize UniProt data across multiple datasets, enabling download, analysis, and extraction of valuable research insights. This website is free and open to all users, and there is no login requirement.

Graphical Abstract

Figure 1.

UniProt website API service architecture: This high-level architecture diagram shows a system where different clients (web, mobile, and programmatic) interact with back-end services via an API Gateway. (a) The API Gateway routes requests to (b) multiple REST services (UniProtKB, UniRef, UniParc, etc.), each running multiple instances for scalability and redundancy. These REST services retrieve data from two back-end storage systems: (c) Search Cluster (Solr)—handles querying of documents for fast search based on user’s search criteria. (d) Key-Value Store Cluster (Voldemort)—provides efficient, low-latency storage for key-value-based lookups.