Ordering permutation for query relative to target

Computes a permutation which rearranges the sequence levels of the query genome so that it reflects the order in which we see the matches on a pairwise plot.

Usage

orderQuerySeqLevels(gb)

orderQuerySeqLevels_DF_GR(DF, gr)

Arguments

gb

A GBreaks object that has only one sequence level in use on the target genome.

DF

A DataFrame object representing the target genome.

gr

A GRanges object representing the query genome.

Value

Returns an integer vector of order permutations for the sequence levels of the query genome.

Details

This is done by computing the average midpoint position of the query ranges on the target genome for each seqlevel, matched by the width of the target ranges, so that long matches have more importance. This avoids spurious ordering due to short matches in the subtelomeric regions.

Note

The order only makes sense relative to a single sequence level of the target genome, so the function will stop with error if there was more than one.

Author

Charles Plessy

Examples

gb       <- GRanges(c("chr1:101-180:+", "chr1:201-300:+",  "chr1:320-400:+"))
gb$query <- GRanges(c( "cgt8:1-100",      "ctg3:1-100",    "ctg5:1-100"))
gb <- GBreaks(gb)
# Sort alphabetically as if they were inherited from a BSgenome object
seqlevels(gb$query) <- c("ctg3", "ctg5", "cgt8")
seqlevels(gb$query)
#> [1] "ctg3" "ctg5" "cgt8"
# Sort by match positions on the target genome.
seqlevels(gb$query) <- seqlevels(gb$query)[orderQuerySeqLevels(gb)]
seqlevels(gb$query)
#> [1] "cgt8" "ctg3" "ctg5"