Hi Francesca,
I'm no missing data expert, but heuristically, I think the
pamr.knnimpute() function makes sense. However, I would look closely
at
your results to make sure you aren't getting too many significant
genes
with imputed values, where 'too many' is a term meant to let me say
something without really saying anything ;-D
Seriously, though, I wouldn't imagine doing the knnimpute would cause
a
bunch of your genes to suddenly look significant for copa when they
wouldn't if you didn't need to impute. I would just recommend the
sanity
check of looking at the results and making sure that the imputation
isn't doing something unexpected.
Best,
Jim
Francesca Guana wrote:
> Dear James,
>
> now the package COPA works well, if the data don't have missing
values. Is
> that right?
> My original data have missing values, so I run pamr.knnimpute to
impute
> missing expression values. Do you think this is a good procedure?
Otherwise
> do you have any suggestion?
>
> Best regards
> Francesca Guana
>
--
James W. MacDonald, M.S.
Biostatistician
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623
**********************************************************
Electronic Mail is not secure, may not be read every day, and should
not be used for urgent or sensitive issues.
