plot.qb.scantwo {qtlbim} | R Documentation |
Plots joint LOD for chromosomes on a two dimensional grid.
## S3 method for class 'qb.scantwo': plot(x, chr, smooth = 3, main, offset, nodiag, slice = NULL, show.locus = TRUE, weight = c("sqrt","count","none"), verbose = FALSE, split.chr, ...)
x |
An object of class qb.scantwo . |
chr |
Vector of chromosomes to plot. Must be integer. |
smooth |
Perform smoothing if > 0 using weighted average over
smooth adjacent points. |
main |
Main title. |
offset |
Offset to make all values non-negative (see below). |
nodiag |
If TRUE do not include diagonal in plot. |
slice |
Take 1-D slice through 2-D surface is not NULL
(see below). |
show.locus |
If a slice , show locus estimate if TRUE . |
weight |
Weights to use for nearest neighbor
smoothing. sqrt is square root of count per locus. Used only
if smooth > 0. |
verbose |
Give verbose feedback if TRUE . |
split.chr |
Split summary by multiple QTL per chromosome (see
details for plot.qb.scanone ). |
... |
Other parameters passed to generic plot function. |
The offset
is used only if \line{qb.scantwo}
used
type
= "estimate" to make values for plotting all
non-negative. Values are rescaled by offset
so that the origin is
at 1 and, by default, the min and max are at 0 and 2, respectively, for
each half of the plot. We need this at this time because
plot.scantwo
does not allow negative values.
The plot.scantwo
argument nodiag
is set to ensure values are all shown and not modified by
plot.scantwo
. Plots with different values for
nodiag
or lower
than the defauls may be non-sensical. For
instance, passing lower = "cond-int"
produces much white area
on the image.
A non-null slice
yields a 1-D view of the 2-D surface. The plots
for slices use plot.scantwo
. The elements of the slice
vector are:
chr
= Chromosome number to slice on;
upper
= Focus on upper triangle of 2-D if TRUE
;
start
= Start position in chromosome chr
(default = 0);
end
= End position in chromosome chr
(default = end of
chromosome);
weight
= Type of weighted mean across chr
by number of MCMC
samples: 0 = unweigthed; 1 = uniform weighting; 2 = position-specific
weighting (default).
The scantwo object being plotted.
Brian S. Yandell, yandell@stat.wisc.edu
example(qb.scantwo)