Directory and Streaming API


The Global Alliance Data Working Group Directory and streaming API Task Team's goal is to bridge the gap between existing local file based access and processing of sequencing reads, and remote web based API access. We are developing a protocol for bulk streaming of read sequencing data, with an initial focus on using existing next-generation sequencing file formats (SAM/BAM/CRAM) and standard internet protocols for transfer, with a future path to others.
Key design principles include:

  • Multiple server implementations encouraged, including file serving and other transcoding based implementations
  • Arbitrary region based slicing and retrieval sequencing data, from single loci over many samples to whole chromosomes on few samples
  • Initial emphasis on using existing widely adopted binary read data formats for transfer
  • Clients should include both command line and genome browsers