[BioC] ERRORS IN LIWONG EXPRESSION VALUES.

[James MacDonald]

I think your function call is correct. Unfortunately, the implementation
of Li and Wong's invariantset method in affy sometimes doesn't converge
due to outliers (as mentioned in the warnings). I don't think there is
anything you can do about it except:

a.) Give up on liwong and use rma, which is a better method anyway

b.) Use dChip, which doesn't have the convergence problems.

Jim



James W. MacDonald
Affymetrix and cDNA Microarray Core
University of Michigan Cancer Center
1500 E. Medical Center Drive
7410 CCGC
Ann Arbor MI 48109
734-647-5623

>>> Stephen Nyangoma <S.Nyangoma at cs.rug.nl> 08/21/03 07:51AM >>>
Dear All, 
I want to obtain expression values based on my data using Li-Wong
method. I gave the following command 

liwongexpre<-expresso(Data2,normalize.method="invariantset",bg.correct=FALSE,
pmcorrect.method="pmonly",summary.method="liwong")

It looks like I do not attain convergence. Is may call function ok?


> Data2
AffyBatch object
size of arrays=640x640 features (12803 kb)
cdf=MG_U74Av2 (12488 affyids)
number of samples=4
number of genes=12488
annotation=mgu74av2
>


Data2<-ReadAffy(filenames=c("../M7SS02080619.CEL","../M7SS02080607.CEL")

> colnames(exprs(Data2))<- c("BXD5a","BXD9a")
>
liwongexpre<-expresso(Data2,normalize.method="invariantset",bg.correct=FALSE,pmcorrect.method="pmonly",summary.method="liwong")
normalization: invariantset
PM/MM correction : pmonly
expression values: liwong
normalizing...done.
12488 ids to be processed
.........
There were 50 or more warnings (use warnings() to see the first 50)


> warnings()
Warning messages:
1: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
2: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
3: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
4: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
5: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
6: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
7: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
8: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
9: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
10: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
11: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
12: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
13: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
14: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
15: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)
16: No convergence achieved, too many outliers in: fit.li.wong(probes,
...)

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch 
https://www.stat.math.ethz.ch/mailman/listinfo/bioconductor