Tissue Specificity R-package

Question

[BioinfGuru]

Hello everyone,

I have written a pipeline that takes count data and calculates tissue specificity. I am planning to make an R-package from this and go for publication. Before I begin packaging and writing drafts, I am doing my literature review and also looking to see if someone has created a package with this function already. Is anyone aware of another package/program that calculates tissue specificity? I certainly couldn't find one - maybe someone else has. Any pointers or advice would be a big help.

Edit: So far:

1) scMCA - works with single cell RNA-seq data

2) Cellmix - works with mircroarray data - is it appropriate to use this for RNA-seq? or on mixed microarray/RNA-seq data? not sure

Thanks,

Kenneth

[GenoMax]

Just saw this publication which includes a new R package called scMCA to do cell type identification.

[BioinfGuru]

Thanks genomax - I'll be looking in to this, however it appears to be for single cell data - i'm guessing (hoping) it wont be able to handle the heterogeneity of non-sc RNA-seq

[BioinfGuru]

Do you have experience of using scMCA or CellMix? advantages/dsadvantages etc.

Answer 1

[Kevin Blighe]

This is an increasing area of interest and current methods are based on cDNA microarray signatures in whole blood that aim to determine concentrations of immune cells, a process generally referred to as cellular deconvolution.

One of the most widely known packages in this regard is CellMix: CellMix: a comprehensive toolbox for gene expression deconvolution.

The main signature that CellMix uses to deconvolute expression data is that identified by Abbas and colleagues in 2009: Deconvolution of blood microarray data identifies cellular activation patterns in systemic lupus erythematosus.

--------------

I have not seen other methods that can do this for RNA-seq data but, coincidentally, I'm currently on a large publication that's under peer review where, as part of the work, we provide a deconvolution process using RNA-seq normalised counts. The work is definitively medical in focus and less so informatics.

Kevin

[BioinfGuru]

That's great help Kevin thank you.

[BioinfGuru]

Same question as for genomax above Kevin, Do you have experience of using scMCA or CellMix? advantages/dsadvantages etc.

[Kevin Blighe]

I have only used CellMix. The main issue is in linking your own study's annotation to that used by CellMix. As annotation conversions is tricky in itself for different reasons, this is a step for which you should devote time and care. CellMix should also only be used for microarray data, which is the type of data on which the signatures are based.

Generally, I find it not a robust method due to the fact that results can vary widely by tweaking a few parameters. There is a need for more robust methods that can work on RNA-seq data.

Granted, it has been slow to develop these tools because we are only now gaining great insight into tissue / cell specific gene expression patterns.