genomesyncr: UCSC Genome Browser Data Synchronizer for R

Goals

• Provide reproducible interface to access most commonly used sequence and annotation data from UCSC
• Download and sync table flatfiles from UCSC's FTP
• Keep UCSC's folder structure: A partial local mirror of the tables you need
• Provide some faculty for parsing the tables based on their schemas and creating GenomicRanges objects from them
• Provide some common joins on KnownGene to add gene symbol, cytoBand, etc (may be duplicating GenomicFeatures)
• Add functions to give common queries based on a gene name (list of exons, list of introns, list of repeats, etc)
• Add functions to give common annotation for a list of any genomic ranges (3' UTR overlap, names of overlapping genes, names of upstream genes, names of downstream genes, etc)

Usage

In development.

Proposed Workflow

• Set directory of goldenPath mirror
• One function to load sequence data into a GRanges object for any species/build
  • Download if not already available
• One function to load annotation data into a GRanges object
  • Check if it exists in our local mirror
  • If it does, check with rsync that it hasn't been updated
  • If updated/never downloaded, pull down new version with rsync
  • Load and parse the table into a GRanges