## ---------------------------------------------------------------------
## A. ON A GAlignments OBJECT
## ---------------------------------------------------------------------
ex1_file <- system.file("extdata", "ex1.bam", package="Rsamtools")
param <- ScanBamParam(what=c("seq", "qual"))
gal <- readGAlignments(ex1_file, param=param)
gal
## This trims 3 nucleotides on the left and 5 nucleotides on the right
## of each alignment:
gal2 <- qnarrow(gal, start=4, end=-6)
gal2
## Note that the 'start' and 'end' values are relative to the query
## sequences and specify the query substring that must be kept for each
## alignment. Negative values are relative to the right end of the query
## sequence.
## Also note that the metadata columns on 'gal' are propagated as-is so
## the "seq" and "qual" matadata columns must be adjusted "by hand" with
## narrow();
mcols(gal2)$seq <- narrow(mcols(gal)$seq, start=4, end=-6)
mcols(gal2)$qual <- narrow(mcols(gal)$qual, start=4, end=-6)
gal2
## Sanity checks:
stopifnot(identical(qwidth(gal2), width(mcols(gal2)$seq)))
stopifnot(identical(qwidth(gal2), width(mcols(gal2)$qual)))
## ---------------------------------------------------------------------
## B. ON A GAlignmentsList OBJECT
## ---------------------------------------------------------------------
gal1 <- GAlignments(
seqnames=Rle(factor(c("chr1", "chr2", "chr1", "chr3")),
c(1, 3, 2, 4)),
pos=1:10, cigar=paste0(10:1, "M"),
strand=Rle(strand(c("-", "+", "*", "+", "-")), c(1, 2, 2, 3, 2)),
names=head(letters, 10), score=1:10)
gal2 <- GAlignments(
seqnames=Rle(factor(c("chr2", "chr4")), c(3, 4)), pos=1:7,
cigar=c("5M", "3M2N3M2N3M", "5M", "10M", "5M1N4M", "8M2N1M", "5M"),
strand=Rle(strand(c("-", "+")), c(4, 3)),
names=tail(letters, 7), score=1:7)
galist <- GAlignmentsList(noGaps=gal1, Gaps=gal2)
galist
qnarrow(galist)
Run the code above in your browser using DataLab