pmapToTranscriptF {ORFik} | R Documentation |
Map range coordinates between features in the transcriptome and genome (reference) space. The length of x must be the same as length of transcripts. Only exception is if x have integer names like (1, 3, 3, 5), so that x[1] maps to 1, x[2] maps to transcript 3 etc.
pmapToTranscriptF( x, transcripts, ignore.strand = FALSE, x.is.sorted = TRUE, tx.is.sorted = TRUE )
x |
GRangesList/GRanges/IRangesList/IRanges to map to transcriptomic coordinates |
transcripts |
a GRangesList/GRanges/IRangesList/IRanges to map against (the genomic coordinates). Must be of lower abstraction level than x. So if x is GRanges, transcripts can not be IRanges etc. |
ignore.strand |
When ignore.strand is TRUE, strand is ignored in
overlaps operations (i.e., all strands are considered "+") and the
strand in the output is '*'. |
x.is.sorted |
if x is a GRangesList object, are "-" strand groups pre-sorted in decreasing order within group, default: TRUE |
tx.is.sorted |
if transcripts is a GRangesList object, are "-" strand groups pre-sorted in decreasing order within group, default: TRUE |
This version tries to fix the shortcommings of GenomicFeature's version. Much faster and uses less memory. Implemented as dynamic program optimized c++ code.
object of same class as input x, names from ranges are kept.
ranges <- IRanges(start = c(5, 6), end = c(10, 10)) seqnames = rep("chr1", 2) strands = rep("-", 2) grl <- split(GRanges(seqnames, IRanges(c(85, 70), c(89, 82)), strands), c(1, 1)) ranges <- split(ranges, c(1,1)) # both should be mapped to transcript 1 pmapFromTranscriptF(ranges, grl, TRUE)