Linear models with different platforms
2
0
Entering edit mode
@sean-davis-490
Last seen 3 months ago
United States
I have a pretty interesting dataset that involves three phenotypes (tumor/normal/cell line) on two different array platforms (quite different). I am interested in fitting a linear model but I do not want to assume that the variance on the two different platforms is the same (it is clearly not). The samples are all biological replicates. Any suggestions? Thanks, Sean
• 907 views
ADD COMMENT
0
Entering edit mode
rgentleman ★ 5.5k
@rgentleman-7725
Last seen 9.6 years ago
United States
have a look at the Bioconductor report series http://www.bepress.com/bioconductor/ (other contributions are welcome) for the paper on the Synthesis of Microarray experiments, which gives some details on fitting mixed effects models (which is sort of what you are asking about). Code etc is available in the GeneMetaEx package at http://www.bioconductor.org/pub/genemetaex best wishes Robert Sean Davis wrote: > I have a pretty interesting dataset that involves three phenotypes > (tumor/normal/cell line) on two different array platforms (quite different). > I am interested in fitting a linear model but I do not want to assume that > the variance on the two different platforms is the same (it is clearly not). > The samples are all biological replicates. Any suggestions? > > Thanks, > Sean > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > -- Robert Gentleman, PhD Program in Computational Biology Division of Public Health Sciences Fred Hutchinson Cancer Research Center 1100 Fairview Ave. N, M2-B876 PO Box 19024 Seattle, Washington 98109-1024 206-667-7700 rgentlem at fhcrc.org
ADD COMMENT
0
Entering edit mode
Fangxin Hong ▴ 810
@fangxin-hong-912
Last seen 10.2 years ago
If you are interested in finding DE (differetially expressed ) genes, you can try RankProd, which combines data from different platforms without assuming equal variance. For linear model without assuming equal variance, it sounds to me that you shouldn't use one linear model since you won't gain any d.f. in estimating error variance. Fangxin > I have a pretty interesting dataset that involves three phenotypes > (tumor/normal/cell line) on two different array platforms (quite > different). > I am interested in fitting a linear model but I do not want to assume that > the variance on the two different platforms is the same (it is clearly > not). > The samples are all biological replicates. Any suggestions? > > Thanks, > Sean > > _______________________________________________ > Bioconductor mailing list > Bioconductor at stat.math.ethz.ch > https://stat.ethz.ch/mailman/listinfo/bioconductor > > -------------------- Fangxin Hong Ph.D. Plant Biology Laboratory The Salk Institute 10010 N. Torrey Pines Rd. La Jolla, CA 92037 E-mail: fhong at salk.edu (Phone): 858-453-4100 ext 1105
ADD COMMENT

Login before adding your answer.

Traffic: 623 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