|
Factor Analysis for Multiple Testing (FAMT) :
an R package for simultaneous tests
under dependence in high-dimensional data
The method proposed in this package takes into account the impact of
dependence on the multiple testing procedures for high-throughput data
as proposed by Friguet et al.
(2009). The common information shared by all the variables is modelised
by a factor analysis structure. First, the number of factors considered
in the model is chosen to reduce the false discoveries variance in
multiple tests. Then, the model parameters are estimated thanks to an
EM algorithm, based on the residual matrix for the fixed part of the
model. Finally, adjusted tests statistics are derived, as well as the
associated p-values.
The proportion of true null hypotheses is estimated, and the Q-values
are calculated, to obtain the rejected hypotheses under each FDR
control level.
Download the package for windows : FAMT_1.2.zip
Short tutorial : FAMT_tutorial.pdf
A poster about the general approach
Reference : Friguet C., Kloareg M. and Causeur D. (2009). A factor model approach to multiple testing under dependence. JASA. In press
|
