BSJ_to_circRNA_sequence: bsj_to_circRNA_sequence

In VCCRI/Ularcirc: Shiny app for canonical and back splicing analysis (i.e. circular and mRNA analysis)

bsj_to_circRNA_sequence

Description

Takes one BSJ coordinate and generates a predicted circular RNA sequence.

Usage

bsj_to_circRNA_sequence(
  BSJ,
  geneID = NULL,
  genome,
  TxDb,
  annotationLibrary,
  reduce_candidates = TRUE,
  shiny = FALSE
)

Arguments

BSJ	: BSJ coordinate in the format of chr_coordinate_chr_coorindate OR chr:coordinate-coorindate:strand.
geneID	: The gene ID that the BSJ aligns to. Not essential as this can be identified from the BSJ coordinate, however time performance of function improved if this information can be provided.
genome	: Is the length f the library fragment
TxDb	: The sequence read length
annotationLibrary	: annotation database. See details for example.
reduce_candidates	: IF multiple exon entries align to a single BSJ then either return longest entry (TRUE) or all entries (FALSE)
shiny	: If TRUE then will setup shiny progress bars. Default is FALSE where a standard text progress bar is used.

Details

Backsplice junction coordinates are typically reported as a character string. Two formats are recognised, ":" delimited (eg circExplorer, CIRI) or "_" delimited (Ularcirc). The BSJ genomic coordinates are compared against the supplied gene model and exonic sequences from matching splice junctions are concatenated. This means the BSJ is the first and last nucleotide of the returned sequence. The current implementation will automatically check 0 or 1 base coordinates and any match is returned.

In some cases one BSJ will match multiple exon combinations. The default setting is to return the longest sequence. Alternatively all possibilities can be returned by setting reduce_candidates to FALSE. BSJ candidates that align to multiple exon combinations are added to duplicated list.

BSJ that do not align to any canonical junctions are returned as failed.

Value

Returns a DNAstring object.

Examples

library('Ularcirc')
TxDb <- TxDb.Hsapiens.UCSC.hg38.knownGene::TxDb.Hsapiens.UCSC.hg38.knownGene
genome <- BSgenome.Hsapiens.UCSC.hg38::BSgenome.Hsapiens.UCSC.hg38
annotationLibrary <- org.Hs.eg.db::org.Hs.eg.db

# Define BSJ. Following two formats are accepted
BSJ <- 'chr2:40430305-40428472:-'       # SLC8A1
BSJ  <- 'chr2_40430305_chr2_40428472'   # SLC8A1

circRNA_sequence <- bsj_to_circRNA_sequence(BSJ, "SLC8A1", genome,TxDb, annotationLibrary)

# You can also retrieve sequence without passing gene annotation - but this is slower
# circRNA_sequence <- bsj_to_circRNA_sequence(BSJ, NULL, genome,TxDb, annotationLibrary)

TxDb <- TxDb.Hsapiens.UCSC.hg38.knownGene::TxDb.Hsapiens.UCSC.hg38.knownGene
genome <- BSgenome.Hsapiens.UCSC.hg38::BSgenome.Hsapiens.UCSC.hg38
# EXAMPLE1 (3 fail and 2 will produce sequences)
BSJ <- c("chr14_99465814_chr14_99458278","chr22_20933778_chr22_20934245",
         "chr12_120155720_chr12_120154969", "chr4_143543508_chr4_143543973",
         "chr10_7285955_chr10_7276891")
GeneIDs <- c("SMARCA5","MSLN","RNF138","KIAA0368","CRKL")
circRNA_sequence <- bsj_to_circRNA_sequence(BSJ, GeneIDs, genome,TxDb, annotationLibrary)

# Returns a list with three items:
# (1) "identified" is a list of DNA strings from BSJ that aligned to FSJ coordinates of the gene model
# (2) "failed" is a character object of BSJ that did not align to FSJ coordinates of gene model. Each entry is
# named with gene ID.
# (3) "duplicates" (not implemented yet) identifies which BSJ returned multiple sequences

VCCRI/Ularcirc documentation built on April 8, 2022, 5:17 p.m.