GenomeInfoDb (version 1.2.4)

fetchExtendedChromInfoFromUCSC: Fetching chromosomes info for some of the UCSC genomes

Description

Fetch the chromosomes info for some of the UCSC genomes. Only supports hg38, hg19, hg18, mm10, mm9, dm3, sacCer3, and sacCer2 at the moment.

Usage

fetchExtendedChromInfoFromUCSC(genome, goldenPath_url="http://hgdownload.cse.ucsc.edu/goldenPath")

Arguments

genome

A single string specifying the UCSC genome e.g. "sacCer3".

goldenPath_url

A single string specifying the URL to the UCSC goldenPath location. This URL is used internally to build the full URL to the 'chromInfo' MySQL dump containing chromosomes information for genome. See Details section below.

Value

A data frame with 1 row per seqlevel in the UCSC genome, and at least 3 columns:

• UCSC_seqlevels: Character vector with no NAs. This is the chrom field of the UCSC 'chromInfo' table for the genome. See Details section above.
• UCSC_seqlengths: Integer vector with no NAs. This is the size field of the UCSC 'chromInfo' table for the genome. See Details section above.
• circular: Logical vector with no NAs. This knowledge is stored in the GenomeInfoDb package itself for the supported genomes.

If the UCSC genome is *not* based on an NCBI assembly (e.g. sacCer2), there are no additional columns and a warning is emitted. Note that, in this case, the rows are not sorted in any particular order.If the UCSC genome is based on an NCBI assembly (e.g. sacCer3), the returned data frame has 3 additional columns:

• NCBI_seqlevels: Character vector. This information is obtained from the NCBI assembly report for the genome. Will contain NAs for UCSC seqlevels with no corresponding NCBI seqlevels (e.g. for chrM in hg18 or chrUextra in dm3), in which case fetchExtendedChromInfoFromUCSC emits a warning.
• SequenceRole: Factor with levels assembled-molecule, unlocalized-scaffold, unplaced-scaffold, alt-scaffold, and pseudo-scaffold. This information is obtained from the NCBI assembly report for the genome. Can contain NAs but no warning is emitted in that case.
• GenBankAccn: Character vector. This information is obtained from the NCBI assembly report for the genome. Can contain NAs but no warning is emitted in that case.

Note that, in this case, the rows are sorted by level in the SequenceRole column, that is, assembled-molecules first, then unlocalized-scaffolds, etc, and NAs last.

Details

Chromosomes information (e.g. names and lengths) for any UCSC genome is stored in the UCSC database in the 'chromInfo' table, and is normally available as a MySQL dump at:

  goldenPath_url//database/chromInfo.txt.gz

fetchExtendedChromInfoFromUCSC downloads and imports that table into a data frame, keeps only the UCSC_seqlevels and UCSC_seqlengths columns (after renaming them), and adds the circular logical column.

Then, if this UCSC genome is based on an NCBI assembly (e.g. hg38 is based on GRCh38), the NCBI seqlevels and GenBank accession numbers are extracted from the NCBI assembly report and the UCSC seqlevels matched to them (using some guided heuristic). Finally the NCBI seqlevels and GenBank accession numbers are added to the returned data frame.

See Also

• The seqlevels getter and setter.

• The seqlevelsStyle getter and setter.

• The getBSgenome utility in the BSgenome package for searching the installed BSgenome data packages.

Examples

## ---------------------------------------------------------------------
## A. BASIC EXAMPLE
## ---------------------------------------------------------------------

## The sacCer3 UCSC genome is based on an NCBI assembly (RefSeq Assembly
## ID is GCF_000146045.2):
sacCer3_chrominfo <- fetchExtendedChromInfoFromUCSC("sacCer3")
sacCer3_chrominfo

## But the sacCer2 UCSC genome is not:
sacCer2_chrominfo <- fetchExtendedChromInfoFromUCSC("sacCer2")
sacCer2_chrominfo

## ---------------------------------------------------------------------
## B. USING fetchExtendedChromInfoFromUCSC() TO PUT UCSC SEQLEVELS ON
##    THE GRCh38 GENOME
## ---------------------------------------------------------------------

## Load the BSgenome.Hsapiens.NCBI.GRCh38 package:
library(BSgenome)
genome <- getBSgenome("GRCh38")  # this loads the
                                 # BSgenome.Hsapiens.NCBI.GRCh38 package

## A quick look at the GRCh38 seqlevels:
length(seqlevels(genome))
head(seqlevels(genome), n=30)

## Fetch the extended chromosomes info for the hg38 genome:
hg38_chrominfo <- fetchExtendedChromInfoFromUCSC("hg38")
dim(hg38_chrominfo)
head(hg38_chrominfo, n=30)

## 2 sanity checks:
##   1. Check the NCBI seqlevels:
stopifnot(setequal(hg38_chrominfo$NCBI_seqlevels, seqlevels(genome)))
##   2. Check that the sequence lengths in 'hg38_chrominfo' (which are
##      coming from the same 'chromInfo' table as the UCSC seqlevels)
##      are the same as in 'genome':
stopifnot(
  identical(hg38_chrominfo$UCSC_seqlengths,
            unname(seqlengths(genome)[hg38_chrominfo$NCBI_seqlevels]))
)

## Extract the hg38 seqlevels and put the GRCh38 seqlevels on it as
## the names:
hg38_seqlevels <- setNames(hg38_chrominfo$UCSC_seqlevels,
                           hg38_chrominfo$NCBI_seqlevels)

## Set the hg38 seqlevels on 'genome':
seqlevels(genome) <- hg38_seqlevels[seqlevels(genome)]
head(seqlevels(genome), n=30)