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.