getENSEMBL.Seq {biomartr}R Documentation

Helper function for retrieving biological sequence files from ENSEMBL

Description

This function downloads gff files of query organisms from ENSEMBL.

Usage

getENSEMBL.Seq(
  organism,
  type = "dna",
  id.type = "toplevel",
  release = NULL,
  path
)

Arguments

organism

Organism selector id, there are three options to characterize an organism:

  • by scientific name: e.g. organism = "Homo sapiens"

  • by database specific accession identifier: e.g. organism = "GCF_000001405.37" (= NCBI RefSeq identifier for Homo sapiens)

  • by taxonomic identifier from NCBI Taxonomy: e.g. organism = "9606" (= taxid of Homo sapiens)

type

character, biological sequence type (e.g. "dna", "cds")

id.type

a character, default "toplevel". id type of assembly, either "toplevel" or "primary_assembly" for genomes. Can be other strings, for non genome objects.

release

a numeric, the database release version of ENSEMBL (db = "ensembl"). Default is release = NULL meaning that the most recent database version is used. release = 75 would for human would give the stable GRCh37 release in ensembl. Value must be > 46, since ensembl did not structure their data if the standard format before that.

path

location where file shall be stored.

Value

either a character path to downloaded file, or a logical FALSE, specifying failure.

Author(s)

Hajk-Georg Drost


[Package biomartr version 1.0.7 Index]