Where did CellMix go? (Anyone know of cell-type deconvolution methods for array data that would work as well?)

Question

[Adamc]

Recently, I was thinking of trying out a cell type deconvolution method ("in silico sorting") to try to remove expression associated with a possibly contaminating cell type in a gene expression profile from samples that were actually FACS sorted but seem to have some expression of markers that are strongly associated with another cell type. I was going to try to use PSEA (which is in BioC now) but I had previously heard a lot about CellMix, so I was thinking of trying that out as well.

Except it looks like the CellMix site is now inaccessible, and has been for at least a month now. I found that I can get it out of https://r-forge.r-project.org/scm/viewvc.php/pkg/CellMix/?root=cellmix but it doesn't have the newest release. Anyone have an archive of CellMix somewhere? Or have any alternative suggestions or approaches to try to achieve a similar result? (I checked archive.org but they didn't capture the whole project site- no downloads, and not all of the documentation either)

Thanks for the assistance.

EDIT: The site for CellMix is back up now.

Answer 1

[Ahill]

For alternative options, you may want to check the link to the Shen-Orr 2013 review (see Table 1 for example) in this post: Differential gene expession analysis in cell populations of mixed tumor and normal cells