multtest error - class label problem
2
0
Entering edit mode
yuhong ning ▴ 20
@yuhong-ning-1763
Last seen 8.1 years ago
Hi, I am using "multtest" to analyze proteomics data. The experiment is to compare protein differential expression between Wild Type and mutant cell types. MS software generates a ratio of intensity of heavy-isotope/light-isotope which correspondent to wild type vs. mutant. It is a triplicate experiment, but for some proteins, ratio data are missing. My R script is very simple: > dd<-read.delim("log2_data.txt", header=TRUE) > clabel<-c(0,0,0) > res<-mt.maxT(dd, clabel, B=20) Error in mt.checkclasslabel(classlabel, test) : in t test, every number in class label needs to be 0 or 1 and neither of the 0 set or 1 set can be empty set The folllowing is your setting of classlabel 0 in t test, every number in class label needs to be 0 or 1 and neither of the 0 set or 1 set can be empty set The folllowing is your setting of classlabel 0 in t test, every number in class label needs to be 0 or 1 and neither of the 0 set or 1 set can be empty set The folllowing is your setting of classlabel 0 Could any one help? Thanks a lot. yuhong
Proteomics Proteomics • 736 views
ADD COMMENT
0
Entering edit mode
Yuhong Ning ▴ 10
@yuhong-ning-1770
Last seen 8.1 years ago
Thank you, Wolfgang. I did use limma to analyze the same data set days ago, and hoped to use "multtest" for a validation of my limma result. So I can't use "multtest" for one group data? Yuhong -----Original Message----- From: bioconductor-bounces@stat.math.ethz.ch [mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Wolfgang Huber Sent: Thursday, June 22, 2006 10:47 AM To: yuhong ning Cc: Katherine S. Pollard; bioconductor at stat.math.ethz.ch Subject: Re: [BioC] multtest error - class label problem Dear Yuhong, the error message says it all, you specify clabel<-c(0,0,0) which corresponds to one group of size three, while mt.maxT expects two groups. Katie - There seems to be a buglet in the code that produces the error message, making it a bit hard to read, could you check this? If you only have three replicates, I would recommend to a moderated t-statistics such as available with the limma package (eBayes function): library(limma) ? eBayes example(eBayes) see also http://www.bepress.com/bioconductor/paper7 Best wishes Wolfgang yuhong ning wrote: > Hi, > > I am using "multtest" to analyze proteomics data. The experiment is to > compare protein differential expression between Wild Type and mutant > cell types. MS software generates a ratio of intensity of > heavy-isotope/light-isotope which correspondent to wild type vs. > mutant. It is a triplicate experiment, but for some proteins, ratio > data are missing. My R script is very simple: > >> dd<-read.delim("log2_data.txt", header=TRUE) >> clabel<-c(0,0,0) >> res<-mt.maxT(dd, clabel, B=20) > Error in mt.checkclasslabel(classlabel, test) : > in t test, every number in class label needs to be 0 or 1 and > neither of the 0 set or 1 set can be empty set The folllowing is your > setting of classlabel 0 in t test, every number in class label needs > to be 0 or 1 and neither of the 0 set or 1 set can be empty set The > folllowing is your setting of classlabel 0 in t test, every number in > class label needs to be 0 or 1 and neither of the 0 set or 1 set can > be empty set The folllowing is your setting of classlabel 0 > > > Could any one help? > > Thanks a lot. > > yuhong > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor -- ------------------------------------------------------------------ Wolfgang Huber EBI/EMBL Cambridge UK http://www.ebi.ac.uk/huber _______________________________________________ Bioconductor mailing list Bioconductor at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/bioconductor Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor ********************************************************* THIS ELECTRONIC MAIL MESSAGE AND ANY ATTACHMENT IS CONFIDENTIAL AND MAY CONTAIN LEGALLY PRIVILEGED INFORMATION INTENDED ONLY FOR THE USE OF THE INDIVIDUAL OR INDIVIDUALS NAMED ABOVE. If the reader is not the intended recipient, or the employee or agent responsible to deliver it to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If you have received this communication in error, please reply to the sender to notify us of the error and delete the original message. Thank You.
ADD COMMENT
0
Entering edit mode
Yuhong, You can use multtest for 1 group data, with the boostrap null distribution (MTP function), but not wiht the permutation nulldistribution. If your sample size is small, however, the bootstrap null distribution will not work. With n=3, it really does not make sense to do resampling based multiple testing. That said, you might try MTP with test="t.onesamp" standarize=FALSE robust=TRUE This will do a 1 sample wilcoxon signed rank test more or less. You might just use wilcox.test also, and forget the resampling. Best, Katie On Thu, 22 Jun 2006, Yuhong Ning wrote: > > > Thank you, Wolfgang. > > I did use limma to analyze the same data set days ago, and hoped to use > "multtest" for a validation of my limma result. So I can't use > "multtest" for one group data? > > Yuhong > > > > > > > -----Original Message----- > From: bioconductor-bounces at stat.math.ethz.ch > [mailto:bioconductor-bounces at stat.math.ethz.ch] On Behalf Of Wolfgang > Huber > Sent: Thursday, June 22, 2006 10:47 AM > To: yuhong ning > Cc: Katherine S. Pollard; bioconductor at stat.math.ethz.ch > Subject: Re: [BioC] multtest error - class label problem > > Dear Yuhong, > > the error message says it all, you specify > clabel<-c(0,0,0) > which corresponds to one group of size three, while mt.maxT expects two > groups. > > Katie - There seems to be a buglet in the code that produces the error > message, making it a bit hard to read, could you check this? > > If you only have three replicates, I would recommend to a moderated > t-statistics such as available with the limma package (eBayes function): > > library(limma) > ? eBayes > example(eBayes) > > see also http://www.bepress.com/bioconductor/paper7 > > > Best wishes > Wolfgang > > > > yuhong ning wrote: >> Hi, >> >> I am using "multtest" to analyze proteomics data. The experiment is to > >> compare protein differential expression between Wild Type and mutant >> cell types. MS software generates a ratio of intensity of >> heavy-isotope/light-isotope which correspondent to wild type vs. >> mutant. It is a triplicate experiment, but for some proteins, ratio >> data are missing. My R script is very simple: >> >>> dd<-read.delim("log2_data.txt", header=TRUE) >>> clabel<-c(0,0,0) >>> res<-mt.maxT(dd, clabel, B=20) >> Error in mt.checkclasslabel(classlabel, test) : >> in t test, every number in class label needs to be 0 or 1 and >> neither of the 0 set or 1 set can be empty set The folllowing is your > >> setting of classlabel 0 in t test, every number in class label needs >> to be 0 or 1 and neither of the 0 set or 1 set can be empty set The >> folllowing is your setting of classlabel 0 in t test, every number in >> class label needs to be 0 or 1 and neither of the 0 set or 1 set can >> be empty set The folllowing is your setting of classlabel 0 >> >> >> Could any one help? >> >> Thanks a lot. >> >> yuhong >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor at stat.math.ethz.ch >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: >> http://news.gmane.org/gmane.science.biology.informatics.conductor > > > -- > ------------------------------------------------------------------ > Wolfgang Huber EBI/EMBL Cambridge UK http://www.ebi.ac.uk/huber > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > ********************************************************* > THIS ELECTRONIC MAIL MESSAGE AND ANY ATTACHMENT IS > CONFIDENTIAL AND MAY CONTAIN LEGALLY PRIVILEGED > INFORMATION INTENDED ONLY FOR THE USE OF THE INDIVIDUAL > OR INDIVIDUALS NAMED ABOVE. > If the reader is not the intended recipient, or the > employee or agent responsible to deliver it to the > intended recipient, you are hereby notified that any > dissemination, distribution or copying of this > communication is strictly prohibited. If you have > received this communication in error, please reply to the > sender to notify us of the error and delete the original > message. Thank You. > ********************************************************* > >
ADD REPLY
0
Entering edit mode
@wolfgang-huber-3550
Last seen 17 days ago
EMBL European Molecular Biology Laborat…
Dear Yuhong, the error message says it all, you specify clabel<-c(0,0,0) which corresponds to one group of size three, while mt.maxT expects two groups. Katie - There seems to be a buglet in the code that produces the error message, making it a bit hard to read, could you check this? If you only have three replicates, I would recommend to a moderated t-statistics such as available with the limma package (eBayes function): library(limma) ? eBayes example(eBayes) see also http://www.bepress.com/bioconductor/paper7 Best wishes Wolfgang yuhong ning wrote: > Hi, > > I am using "multtest" to analyze proteomics data. The > experiment is to compare protein differential > expression between Wild Type and mutant cell types. MS > software generates a ratio of intensity of > heavy-isotope/light-isotope which correspondent to > wild type vs. mutant. It is a triplicate experiment, > but for some proteins, ratio data are missing. My R > script is very simple: > >> dd<-read.delim("log2_data.txt", header=TRUE) >> clabel<-c(0,0,0) >> res<-mt.maxT(dd, clabel, B=20) > Error in mt.checkclasslabel(classlabel, test) : > in t test, every number in class label needs > to be 0 or 1 and neither of the 0 set or 1 set can be > empty set > The folllowing is your setting of classlabel 0 > in t test, every number in class label needs to be 0 > or 1 and neither of the 0 set or 1 set can be empty > set > The folllowing is your setting of classlabel 0 > in t test, every number in class label needs to be 0 > or 1 and neither of the 0 set or 1 set can be empty > set > The folllowing is your setting of classlabel 0 > > > Could any one help? > > Thanks a lot. > > yuhong > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: http://news.gmane.org/gmane.science.biology.informatics.conductor -- ------------------------------------------------------------------ Wolfgang Huber EBI/EMBL Cambridge UK http://www.ebi.ac.uk/huber
ADD COMMENT

Login before adding your answer.

Traffic: 314 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6