Description Usage Arguments Details Value Author(s) References Examples
Estimate the q-values for a given set of p-values, in a
safe manner. The most frequent error that I have seen in
using qvalue
, is that pertaining to the
“pi_0 estimate is less than zero” which is thrown
as a result of not being able to estimate the lambda
parameter. This function first attempts to generate
qvalues using the default “smoother” method. If
that fails due to pi_0 estimate, then this is repeated
using the “bootstrap” method. If the same error is
still thrown, then a fallback FDR estimation method is
used, which defaults to “BH”, the
Benjamini-Hochberg FDR
1 2 |
p |
a vector of p-values |
lambda |
The value of the tuning parameter to estimate pi_0. Must be in [0,1). Optional, see Storey (2002). |
fallback |
The fallback p.adjust method. Default =
“BH”, but can be one of the other FDR methods from
|
In addition, when running library(qvalue)
in a
headless server, you often get a warning: In
fun(libname, pkgname) : no DISPLAY variable so Tk is not
available
which is actually thrown by tcltk. qvalue2
suppresses that message. See references.
An object of class qvalue
, even if the the FDR was
estimated using the fallback
option, where the
latter has pi0=1.0
.
Mark Cowley, 2011-10-25
http://stackoverflow.com/a/11554019/178297
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 | p <- runif(100,0,1)
qvalue2(p)
# qv
pbad <- c( 2e-04, 3e-04, 5e-04, 6e-04, 8e-04, 8e-04, 9e-04, 9e-04, 0.001,
0.001, 0.0013, 0.0013, 0.0013, 0.0014, 0.0016, 0.0017, 0.0019,
0.0019, 0.0023, 0.0023, 0.0023, 0.0024, 0.0028, 0.0029, 0.0031,
0.0032, 0.0032, 0.0032, 0.0034, 0.0034, 0.0035, 0.0037, 0.0038,
0.0038, 0.0042, 0.0043, 0.0044, 0.0044, 0.0044, 0.0045 )
qvalue2(pbad)
# These will now throw errors:
## Not run:
qvalue2(p=c(-0.1,0,0.5,0.9,1.0))
qvalue2(p, lambda=seq(-0.1,1,0.05))
qvalue2(p, lambda=c(0,0.2,0.5))
## End(Not run)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.