Principal component analysis for proteomics data ( like SNPRelate )

Question

[William]

Is there a R package that can be used for doing a pca analysis on proteomics data? To plot which samples are similar to each other.?

Something like what SNPrelate can do for genomics variant data (vcf)? In which kind of format would I need to have the proteomics data?

http://corearray.sourceforge.net/tutorials/SNPRelate/

Answer 1

[Chris Evelo]

I would think that PCA for proteomics is in fact a lot simpler than for SNP data. SNPRelate does a lot of pretreatment steps that are not needed assuming you have proteomics results (as in amounts of individual proteins with some identifier even if that is just a spot location for different conditions. Unless I am wrong that means you can just do basic PCA for R. See e.g. here. There are a number of "easy to use" packages out there. If that is what you are looking for just Google for "Principal Component Analysis in R"

[William]

Why are the pretreatment steps needed for snp data and not proteomics data? I guess my real question is which dimensions are / can / should be used for pca on snp or on proteomics data? Does snp data have less dimensions than proteomics ( protein id / quantification ) data? Is every SNP call or protein identified one dimension?